EA048784B1 - CD1d DOUBLE ACTION IMMUNOGLOBULIN - Google Patents

Abstract

The invention relates to the field of immunology, more specifically to the field of binding molecules that bind to human CD1d and Vy9V62-TCR, including antibodies and fragments thereof that modify CD1d-mediated biological functions, such as stimulating the activation and reducing the activation of CD1d-restricted T cells, including natural killer T cells and gamma-delta T cells, and modulating the function of cells expressing CD1d. The invention also relates to pharmaceuticals and the use of such binding fragments in the treatment of diseases or disorders.

Description

Field to which the invention relates

The present invention relates to the field of immunology, and more particularly to the field of binding molecules and/or immunoglobulins that bind to human CD1d, including antibodies and fragments thereof that modify CD1d-mediated biological functions, such as activation and blocking of CD1d-restricted T cells, including natural killer T cells (NKT) and gamma(',')-delta(<5) T cells, and modulation of the function of cells expressing CD1d. The invention also relates to bi-, tri-, or multispecific binding molecules and/or immunoglobulins that bind to CD1d and the γδ T cell receptor (TCR) and/or a target tumor. The invention also relates to pharmaceuticals and the use of such mono-, bi-, tri-, or multispecific binding molecules and/or immunoglobulins in the treatment of diseases or disorders.

State of the art

CD1d is a member of the CD1 (cluster of differentiation 1) family of glycoproteins (including CD1a, CD1b, CD1c, CD1d, and CD1e) and is expressed on the surface of a variety of human cells, including antigen-presenting cells (APCs). They are non-classical MHC proteins related to MHC class I proteins and are involved in the presentation of lipid antigens to a subset of T cells. In humans, CD1d is encoded by CD1D, also known as R3G1. APCs with CD1d include B cells, dendritic cells (eg, in lymph nodes), and monocytes. CD1d is also expressed by a variety of other cell types, such as the liver, pancreas, skin, kidney, uterus, conjunctiva, epididymis, thymus, and tonsil (Canchis et al. (1992) Immunology 80:561).

Cells that are activated/stimulated via CD1d include CD1d-restricted natural killer T cells (NKT cells). NKT cells are a diverse group of T cells that have properties of both T cells and natural killer cells. NKT cells are a subset of T cells that express the alpha(a)/beta(b) T cell receptor (TCR) as well as a number of molecular markers that are typically associated with NK cells.

Type I NKT or invariant NKT cells (iNKT) are the best-studied group of NKT cells, which differ from standard αβ T cells in that their T cell receptors are much more restricted in diversity (invariant). These invariant, but also other CD1d-restricted NKT cells (type II NKT), recognize several antigens such as (self or foreign) lipids, glycolipids, sulfatides, phospholipids, lipopeptides, hydrophobic peptides and/or amphipathic α-helical peptides presented by CD1d molecules present on APCs (Enrico Girardi et al. (2016) J Biol Chem. 291(20): 10677). The interaction between (antigen-presenting) CD1d and TCR triggers the release of cytokines, including Th1- and/or ThZ-like cytokines such as interferon-γ, tumor necrosis factor-α, and interleukins such as IL-4, IL-5, and IL-13.

α-Galactosylceramide KRN7000 is the best-studied ligand for lipid-binding CD1d in NKT cell activation in vitro and in vivo. Other ligands include isoglobotrigexosylceramide (for murine iNKT cells), (microbial-derived) glycuronosylceramides, α-C-galactosylceramides, threotolceramide, and a number of (human and non-human) glycolipids such as lysophosphatidylcholine and lysosphingomyelin (Fox et al. (2009) PLOS Biology 7:10:e1000228).

An important role for iNKT cells has been demonstrated in regulating autoimmune, allergic, antimicrobial, and antitumor immune responses (van der Vliet et al. (2004) Clin Immunol 112(1): 8). Physiologically, NKT cells can complement or inhibit immune responses, including antitumor, autoimmune, and antipathogen responses, through a number of context-dependent mechanisms (Yue et al. (2010) J Immunol 184: 268), including inducing cell death in the case of multiple myeloma cells. Conditions in which (invariant) NKT cells may be involved include autoimmune or inflammatory diseases, including myasthenia gravis, psoriasis, ulcerative colitis, primary biliary cirrhosis, colitis, autoimmune hepatitis, atherosclerosis, and asthma. In addition to cytokine release, NKT cell effector functions that lead to cell lysis, such as perforin release and granzyme release and cell death, may also be relevant in settings in which NKT cells are involved, such as cancer.

CD1d-restricted NKT cells have been divided into two main subsets based on their T cell receptor (TCR) usage and antigen specificity: type I NKT cells, which use the standard invariant TCR α-chain and recognize α-galactosylceramide (α-GalCer), and type II NKT cells, which use a more diverse αβ-TCR repertoire and do not recognize α-GalCer. In addition, α-GalCer-reactive NKT cells, which use atypical αβ-TCRs, and CD1d-restricted T cells, which use γδ- or δ/αβ-TCRs, have recently been discovered, increasing the diversity of CD1d-restricted T cells (Macho-Fernandez and Brigl (2015) Front Immunol 6:362).

- 1 048784

NKT cells type I.

The discovery of the first CD1d-presenting antigen, α-galactosylceramide (α-GalCer), by Kawano and colleagues in 1997 was an important step toward our understanding of NKT cell biology, particularly type I NKT cells or invariant NKT cells (iNKT) (Kawano et al. (1997) Science 278:1626). Type I NKT cells express the invariant Va14Ja18 TCR α-chain in mice and Va24Ja18 in humans, associated with a limited repertoire of TCR β-chains (Ve8, Ve7, Ve2 in mice and exclusively Ve11 in humans). Type I NKT cells can be highly autoreactive even when quiescent and exhibit an activation/memory phenotype with high surface levels of the activation markers CD69, CD44, and CD122 (IL-2R β chain) and low expression of CD62L, a marker expressed by naive T cells that reside in lymph nodes (Bendelac et al. (1992) J Exp Med 175:731; Matsuda et al. (2000) J Exp Med 192:741). Type I NKT cells play a critical role in natural antitumor responses, as demonstrated by the rapid growth of spontaneous tumors in type I NKT cell-deficient Ja18 ^-/- mice compared with wild-type mice (Smyth et al. (2000) J Exp Med 191:661; Swann et al. (2009) Blood 113:6382; Bellone et al. (2010) PLoS One 5:e8646). Moreover, activation of type I NKT cells by α-GalCer provides high efficacy against hematological malignancies and solid tumors due to their production of IFN-γ and subsequent activation of dendritic cells (DCs) and NKT cells (Smyth et al. (2002) Blood 99:1259; Berzofsky and Terabe (2008) J Immunol 180:3627). In contrast, sulfatide-activated type II NKT cells suppress antitumor immunity (Terabe et al. (2000) Nat Immunol 1:515; Terabe et al. (2005) J Exp Med 202:1627; Renukaradhya et al. (2008) Blood 111:5637) by abrogating type I NKT activation in response to α-GalCer in the context of cytokine secretion and expansion (Ambrosino et al. (2007) J Immunol 179:5126). Moreover, their production of IL-13 in combination with TNF-α leads to an increase in the level of TGF-β secretion by myeloid-derived suppressor cells (MDSC) and results in a decrease in the activity of cytotoxic T cells (Terabe et al. (2003) J Exp Med 198:1741).

NKT cells type II.

CD1d-restricted T cells, which do not express the Va14-Ja18 band and do not recognize α-GalCer, were first described in MHC II-deficient mice among other CD4+ T cells (Cardell et al. (1995) J Exp Med 182:993). Among NKT cells, then called diverse (dNKT), type II NKT, or variant NKT (vNKT), this group of NKT cells, present in both mice and humans, exhibits a more heterogeneous TCR repertoire. Several lines of evidence suggest that type II NKT cells may contribute to and modulate a variety of immune responses, sometimes playing roles opposing those of type I NKT cells.

γδ T cells.

In addition to CD1d-restricted T cells with αβ TCR, CD1d-restricted T cells expressing γδ TCR have recently been described in mice and humans. Based on their Vδ chain expression, human γδ T cells can be divided into two main groups: Vδ2+ and non-Vδ2 subsets, with the latter including, inter alia, Vδ1+γδ T cells and the less common Vδ3+γδ T cells (McVay et al. (1999) Crit Rev Immunol 19:431; Vantourout and Hayday (2013) Nat Rev Immunol 13:88). Vδ1+γδ T cells are primarily tissue-resident and reside in the skin and mucosa, whereas Vδ2+γδ T cells reside in human blood. Compared with αβ T cells, the types of antigens recognized by γδ T cells and the role and function of antigen presentation in γδ TCR recognition are much less well understood. Interestingly, some γδ T cells have recently been discovered that directly recognize CD1d-presented lipid antigens (Hayday and Vantourout (2013) Immunity 39:994). In 2013, Uldrich et al. (2013) Nature Immunol 14: 1137) identified a subset of γδ T cells that recognize CD1d in combination with a range of glycolipid antigens, including αGalCer. These T cells are referred to as CD1d-restricted Vδ1+ T cells. Without being bound by theory, it is believed that Vδ1+ T cells may elicit an inflammatory response that may be counterproductive to an effective antitumor response.

Brief description of the invention

For the above reasons, modulation of CD1d-mediated effects is associated with potential therapeutic benefit. There is a continuing need for binding molecules that can specifically bind and/or interact with CD1d, both in vitro and in vivo. In particular, there is a need for such binding molecules that bind and/or modulate (activate or inhibit) biological functions that are associated with CD1d, such as, but not limited to, NKT cell activation. Such binding molecules may, for example, be useful in various diseases in which CD1d-mediated functions play a role.

The authors previously identified a single-domain antibody, designated 1D12 or VHH 12, that specifically activates iNKT cells (NKT type I) without the need for CD1d ligand (WO2016122320). The present invention shows for the first time that 1D12 is not only capable of activating

- 2 048784 to activate iNKT cells, but at the same time provides blockade of activation of CD1d-restricted Vδ1+ T cells. This observation now leads to the conclusion that 1D12 may be used to treat diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells. One skilled in the art will recognize that this specific use is not limited to 1D12, but also encompasses any antibody that is functionally and/or structurally similar to 1D12.

In a first aspect, the invention provides a binding molecule comprising a first binding moiety that is capable of competing with the single domain antibody 1D12 (SEQ ID NO: 4) for binding to a CD1d molecule, for use in the treatment of diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells.

In a second aspect, the invention provides a binding molecule comprising a first binding moiety that is capable of competing with 1D12 (SEQ ID NO: 4) for binding to a CD1d molecule and comprising a second binding moiety that is capable of specifically binding to a Vγ9Vδ2 TCR, wherein the binding molecule is capable of activating Vγ9Vδ2 T cells. The authors have shown that a bispecific CD1d/Vγ9Vδ2 TCR antibody can induce degranulation of iNKT cells and Vγ9Vδ2 T cells, and control the growth of CD1d-expressing tumor cells. Moreover, in a mouse multiple myeloma model, infusion of both iNKT cells and γδ T cells with the bispecific CD1d/Vγ9Vδ2-TCR antibody significantly prolonged survival compared to a mixture of iNKT cells and γδ T cells alone without this antibody.

In another aspect, the invention provides compositions according to the second aspect of the invention and a pharmaceutically acceptable excipient.

In other aspects, the invention provides a method of treating a disease or disorder comprising administering a binding molecule according to the invention to a subject in need thereof.

Other aspects and embodiments of the invention are described below.

Description of figures

Fig. 1. Anti-CD1d VHH-mediated activation and inhibition of invariant natural killer T (iNKT) cells. iNKT CD25 expression and interferon-γ (IFN-γ) production were determined after co-culture of iNKT cells with CD1d-transfected HeLa cells for 24 h in a pulsed mode with vehicle control (vehicle) or aGalCer with/without combination with anti-CD1d VHH 1D12 (100 nM) or 1D22 (1000 nM). Corresponding dot plots illustrating marked activation (A) or inhibition (B) of iNKT cells by 1D12 are shown with CD25 expression indicated, along with marked increases (by 1D12) or decreases (by 1D22) in iNKT cell IFN-γ production (C and D). Data represent mean+SD of 3-4 separate experiments with iNKT cells from different donors, **P<0.01, ***P<0.001, calculated by one-way ANOVA with Tukey's post hoc test.

Fig. 2 - Vδ1-sulfatide-CD1d interaction is prevented by anti-CD1d VHH 1D12 and reduced by 1D22. Endogenous or sulfatide-loaded CD1d-PE tetramers were incubated with PBS (control), 1D12, or 1D22 (VHH:CD1d ratio 4:1), and the tetramers were used to stain Jurkat-Vδ1 cells (final tetramer concentration 2 μg/ml) in combination with CD3-APC. Tetramer binding was prevented by 1D12 and reduced by 1D22 (A). C1R-CD1d cells were incubated with DMSO or sulfatide controls (25 μg/ml), followed by the addition of medium or anti-CD1d VHH at 1000 or 100 nM and incubation for 30 min. Jurkat-Vδ 1 cells were then added and the co-culture was incubated for an additional 24 h, after which CD69 expression was determined by flow cytometry (B). N=3.

Fig. 3. Dual activation of iNKT cells and Vγ9Vδ2 T cells by a bispecific CD1d- and Vγ9Vδ2-targeted TCR antibody VHH results in degranulation of effector cells and rapid lysis of tumor cells. CD1d-expressing CCRF-CEM cells were incubated with iNKT cells, Vγ9Vδ2 T cells, or both at an effector:target ratio of 1:2 in the presence of medium alone, monovalent 1D12, or bispecific 1D12-5C8. Rapid degranulation (as demonstrated by CD107a expression) of iNKT cells was observed in the presence of 1D12, but only simultaneous degranulation of iNKT cells and Vγ9Vδ2 T cells was observed in the presence of bispecific VHH (A). Accordingly, a striking reduction in the number of viable CCRF-CEM cells was observed (B). Data represent mean+SD of 3 separate experiments with iNKT/ Vγ9Vδ2-T cells obtained from different donors.

Fig. 4. Bispecific 1D12-5C8 antibody supports iNKT and Vγ9Vδ2 T cell expansion and induces tumor growth control. iNKT, Vγ9Vδ2-T, or their mixture (at a 2:3 ratio) were incubated with MM.1s-CD1d cells at an effector:target ratio of 1:10 in the presence of medium alone or bispecific 1D12-5C8. A clear induction of iNKT cell expansion was observed, but Vγ9Vδ2-T expansion was observed only in the presence of 1D12-5C8 and iNKT cells (A). It is noteworthy that tumor growth was

- 3,048,784 cells were inhibited in the presence of 1D12-5C8 (B). Data represent the mean+SD of 3 separate experiments with coupled iNKT/Vγ9Vδ2-T obtained from different donors.

Fig. 5. Anti-CD1d VHH 1D12 and anti-CD1d 51.1 mAb, but not anti-CD1d VHH 1D22, interfere with binding of 1D12-5C8 bispecific VHH. CD1d-transfected multiple myeloma cells (MM.1s) were incubated for 45 min with PBS (negative control, NC), anti-CD1d VHH (1000 nM), or anti-CD1d mAb (100 nM), followed by addition of PBS (NC) or biotinylated 1D125C8 bispecific VHH antibody (100 nM) and incubation for an additional 30 min. After extensive washing, samples were stained with streptavidin-APC and analyzed by flow cytometry. Data represent mean+SD of 3 separate experiments, ****P<0.0001, calculated using two-way ANOVA with Tukey's post hoc test.

Fig. 6 - Bispecific antibody 1D12-5C8 promotes survival in a murine multiple myeloma model in the presence of iNKT and/or γδ T cells. Fig. 6A shows the effects of administration of antibody 1D12-5C8 and/or iNKT cells on survival. Fig. 6B shows the effects of administration of antibody 1D12-5C8 and/or γδ T cells. Fig. 6C shows the effects of administration of antibody 1D12-5C8 and/or a mixture (mix) of iNKT and γδ T cells.

Detailed description of the invention

As described above, in a first main aspect, the invention provides a binding molecule comprising a first binding moiety that is capable of competing with the single domain antibody 1D12 (SEQ ID NO: 4) for binding to a CD1d molecule, for use in the treatment of diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells.

A disease caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells is a disease or disorder in which activation of CD1d-restricted Vδ1+ T cells initiates, maintains or worsens the disease or disorder, or has a negative impact on the prognosis or progression of the disease or disorder. A negative impact is understood to mean that the disease is maintained, worsened or attenuated to a lesser extent by the action of CD1d-restricted Vδ1+ T cells compared to the situation in the absence of Vδ1+ T cell action. Preferably, the negative impact is related to the (anti-)inflammatory effects of activated Vδ1+ T cells. Such disorders or diseases include in particular those that require a type 1 helper T cell (Th1) response and that are negatively affected by a type 2 helper T cell (Th2) response. The designations Th1 and Th2 cells are well known in the art and it is generally accepted that Th1 cells are predominantly active against intracellular microbiota such as intracellular bacteria and viruses, while Th2 cells predominantly activate the humoral immune system (B cells, eosinophils, etc.) to combat extracellular microbiota such as protozoa. Furthermore, Th1 cells promote tumor cell death while Th2 cells counteract this.

One example of a disorder caused, maintained, and/or promoted by activation of CD1d-restricted Vδ1+ T cells is peripheral CD1d-restricted T-cell lymphoma (PTCL (Bachy et al. (2016) J Exp Med 213 No. 5: 841)). In a preferred embodiment of the invention, there is provided a binding molecule for use according to the invention for use in the treatment of hematological malignancies such as peripheral CD1d-restricted T-cell lymphoma (PTCL), CD1d+ T-cell acute lymphoblastic leukemia (ALL), CD1d+ lymphoma, CD1d+ chronic lymphocytic leukemia (CLL), CD1d+ acute myelogenous leukemia (AML) and CD1d+ mantle cell lymphoma, preferably for use in the treatment of CD1d-restricted Vδ1+ PTCL.

The binding molecule according to the invention may be any kind of binding molecule, such as a complex, provided that it comprises or consists of a first binding moiety as defined in the invention. Preferably, the binding molecule is a polypeptide, more preferably an antibody. In some embodiments, the binding molecule may consist only of a first binding moiety that is capable of competing with 1D12 (SEQ ID NO: 4) for binding to human CD1d. In other embodiments, the binding molecule may consist of a first binding moiety that is capable of competing with 1D12 for binding to human CD1d and a label. In other embodiments, the binding molecule may comprise a first binding moiety that is capable of competing with 1D12 for binding to human CD1d, linked to a pharmaceutically active agent and/or another binding moiety(s). Thus, the binding molecule may comprise one binding moiety or several, such as two, three or four binding moieties. In a preferred embodiment of the invention, the binding molecule comprises two binding moieties, preferably capable of binding to two different epitopes. Preferably, the epitopes are located on different targets (proteins).

- 4 048784

The term "antibody" as used herein means an immunoglobulin molecule, a fragment of an immunoglobulin molecule, or a derivative thereof that has the ability to specifically bind to an antigen under standard physiological conditions with a significant half-life, such as at least about 30 minutes, at least about 45 minutes, at least about one hour, at least about two hours, at least about four hours, at least about 8 hours, at least about 12 hours, at least about 24 hours or more, at least about 48 hours or more, about 3, 4, 5, 6, 7 or more days, and so on, or for any other appropriate functionally defined period (e.g., a time sufficient to induce, promote, stimulate, and/or modulate a physiological response associated with the antibody binding to the antigen, and/or a time sufficient for the antibody to acquire effector activity). The binding region (or binding domain) that interacts with the antigen includes the variable regions of both the heavy and light chains of the immunoglobulin molecule. The constant regions of antibodies (Abs) can mediate the binding of immunoglobulin to tissues or host factors, including various cells of the immune system (such as effector cells and T cells) and components of the complement system, such as C1q, which is the first component of the classical complement pathway. In some embodiments, however, the Fc region of the antibody has been modified to be inert, wherein inert is meant an Fc region that is at least incapable of binding any FcY receptors, inducing Fc-mediated cross-linking of FcRs, or inducing Fcr-mediated cross-linking of target antigens by two Fc regions of individual proteins, such as antibodies. In another embodiment, the inert Fc region is further incapable of binding C1q. In one embodiment, the antibody comprises mutations at positions 234 and 235 (Canfield and Morrison (1991) J Exp Med 173:1483), such as a Leu-Phe mutation at position 234 and a Leu-Glu mutation at position 235. In another embodiment, the antibody comprises a LeuAla mutation at position 234, a Leu-Ala mutation at position 236, and a Pro-Gly mutation at position 329.

As indicated above, as used herein, the term antibody, unless otherwise indicated or clearly apparent from the context, includes fragments of an antibody that retain the ability to specifically interact, such as bind, an antigen. It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Examples of binding fragments encompassed by the term antibody include: (i) a Fab' or Fab fragment, a monovalent fragment consisting of the VL, VH, CL and CH1 domains, or a monovalent antibody as described in WO2007059782; (ii) an F(ab') ₂ fragment, bivalent fragments comprising two Fab fragments linked by a disulfide bridge at the hinge region; (iii) an Fd fragment consisting essentially of the VH and CH1 domains; and (iv) an Fv fragment consisting essentially of the VL and VH domains of one arm of an antibody. Moreover, although the two domains of the Fv fragment, VL and VH, are encoded by different genes, they can be linked by recombinant methods by a synthetic linker that allows them to be produced as a single protein chain in which the VL and VH regions are joined to form monovalent molecules (known as single-chain antibodies or single-chain Fv (scFv); see, e.g., Bird et al., Science 242, 423-426 (1988) and Huston et al., PNAS USA 85, 5879-5883 (1988)). Such single-chain antibodies are encompassed by the term antibody unless otherwise indicated or clearly evident from the context. Although such fragments are generally included within the meaning of the term antibody, they collectively and individually represent unique features of the present invention, exhibiting distinct biological properties and uses. These and other useful antibody fragments in the context of the present invention are described further below. It should also be understood that the term antibody, unless otherwise specified, also includes polyclonal antibodies, monoclonal antibodies (mAbs), chimeric antibodies and humanized antibodies, and antibody fragments that retain the ability to specifically bind to an antigen (antigen-binding fragments) obtained by any known method, such as enzymatic cleavage, peptide synthesis, and recombinant methods. The resulting antibody may be of any isotype.

The term immunoglobulin heavy chain, immunoglobulin heavy chain or heavy chain as used herein refers to one or more immunoglobulin chains. A heavy chain typically consists of a heavy chain variable region (abbreviated herein as VH) and a heavy chain constant region (abbreviated herein as CH), which determine the antibody isotype. The heavy chain constant region typically consists of three domains: CH1, CH2 and CH3. The heavy chain constant region may also include a hinge region. The term immunoglobulin as used herein refers to a class of structurally related glycoproteins consisting of two pairs of polypeptide chains, namely one pair of light (L) chains and one pair of heavy (H) chains, all of which are potentially linked together by disulfide bonds. Within the structure of an immunoglobulin (e.g. IgG), the two heavy chains are linked together by disulfide bonds at what is called the hinge region. Like the heavy chains, each light chain is typically made up of several regions: the variable region of the light chain (abbreviated here as VL) and the constant region of the light chain (abbreviated here as VL).

- 5 048 784 (abbreviated herein as CL). The constant region of the light chain typically consists of a single domain: CL. Furthermore, the VH and VL regions can be further subdivided into regions of hypervariability (or hypervariable regions, which may be hypervariable in sequence and/or form structurally distinct loops), also called complementarity determining regions (CDRs), separated by regions that are more conserved, called framework regions (FRs). Each of the VH and VL typically consists of three CDRs and four FRs, arranged from amino- to carboxyl-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The sequences of the CDRs can be determined by various methods, for example those described by Choitia and Lesk (1987) J. Mol. Biol. 196:901 or Kabat et al. (1991) Sequence of protein of immunological interest, fifth edition. NIH publication. Various methods for defining CDRs and amino acid numbering are described at www.abysis.org (UCL).

The term "isotype" as used herein means an immunoglobulin (sub)class (e.g., IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM) or any allotype thereof, such as IgG1m(za) and IgG1m(f) [SEQ ID NO: 407]), which is encoded by the heavy chain constant region genes. Thus, in one embodiment, the antibody comprises an immunoglobulin heavy chain of the IgG1 class or any allotype thereof. Furthermore, each heavy chain isotype may be combined with a kappa (k) or lambda (λ) heavy chain.

The term chimeric antibody as used herein means an antibody in which the variable region is derived from a non-human species (e.g., derived from a rodent) and the constant region is derived from another species, such as a human. Chimeric antibodies can be produced by modifying antibodies. Antibody modification is a general term used for various types of antibody modifications and is a process well known to those skilled in the art. Thus, a chimeric antibody can be a genetically modified recombinant antibody. Some chimeric antibodies can be genetically or enzymatically modified at the same time. The production of a chimeric antibody is within the skill of the art, so the production of a chimeric antibody according to the present invention can be accomplished by methods other than those described herein. Chimeric monoclonal antibodies for therapeutic use are designed to reduce the immunogenicity of antibodies. They may typically comprise variable regions from a non-human species (e.g., murine) that are specific for the antigen of interest, as well as constant domains from the heavy and light chains of a human antibody. The terms variable region or variable domains in the context of chimeric antibodies mean a region that includes the CDRs and framework regions of both the heavy and light chains of an immunoglobulin. The term humanized antibody, as used herein, means a genetically modified antibody of a non-human species that comprises the constant domains of a human antibody and the variable domains of a non-human species modified to include a high level of sequence homology to the human variable domains. This may be achieved by transferring the six complementarity determining regions (CDRs) of a non-human antibody, which together form the antigen binding site, into a homologous human acceptor framework region (FR). In order to fully recapitulate the binding affinity and specificity of the parent antibody, it may be necessary to replace framework residues of the parent antibody (i.e., the non-human antibody) with human framework regions (backmutations). Structural homology modeling can help identify amino acid residues in the framework regions that are important for the binding properties of the antibody. Thus, a humanized antibody may include non-human CDR sequences, primarily human framework regions, optionally including one or more amino acid backmutations in the non-human amino acid sequence, and complete human constant regions. Optionally, additional amino acid modifications, which are not necessarily backmutations, may be used to obtain a humanized antibody with favorable characteristics, such as affinity and biochemical properties. The amino acid sequence of an antibody derived from a non-human species differs from that of human antibodies, and therefore the non-human antibody is potentially immunogenic when administered to human patients. However, despite the fact that an antibody is derived from a non-human species, its CDR fragments are responsible for the ability of the antibody to bind to its target antigen, and the goal of humanization is to preserve the specificity and binding affinity of the antibody. Thus, humanization of therapeutic antibodies from non-human species is carried out to minimize their immunogenicity in humans, while at the same time such humanized antibodies retain the specificity and binding affinity of the antibody from the non-human species.

In one aspect, the present invention provides a multispecific antibody comprising at least a first binding fragment of a binding molecule according to any aspect or embodiment described herein, and one or more binding

- 6,048,784 fragments that bind one or more other targets than the first binding site. Such a multispecific antibody may be a bispecific, trispecific, or tetraspecific antibody.

The term multispecific antibody means an antibody having specificity for at least two different, for example at least three, usually non-coinciding epitopes. Such epitopes may be located on the same target or on different targets. If the epitopes are located on different targets, such targets may be located on the same cell or on different cells or on different cell types.

The term "bispecific antibody" means an antibody having specificity for at least two different, typically non-coincident epitopes. Such epitopes may be located on the same target or on different targets. If the epitopes are located on different targets, such targets may be located on the same cell or on different cells or on different cell types. In one embodiment, the bispecific antibody includes a first and a second heavy chain.

Examples of bispecific antibody molecules that can be used in the present invention include: (i) a single antibody that has two arms comprising different antigen-binding sites, (ii) a single chain antibody that has specificity for two different epitopes, such as two scFvs linked together by an additional peptide linker; (iii) a dual variable domain antibody (DVD-IgTM), in which each light chain and heavy chain comprises two variable domains linked by a short peptide bond; (iv) a chemically linked bispecific (Fab')2 fragment; (v) TandAb®, which is a fusion product of two single chain diabodies, resulting in a tetravalent bispecific antibody with two binding sites for each of the target antigens; (vi) a flexibody, which is a combination of an scFv with a diabody, resulting in a multivalent molecule; (vii) the so-called dock and lock molecule (Dock-and-Lock®) based on the dimerization and docking domain of protein kinase A, which when used with Fab can provide a trivalent bispecific binding protein consisting of two identical Fab fragments linked to another Fab fragment; (viii) the so-called Scorpion molecule, comprising, for example, two scFvs fused to the two ends of a human Fab arm; and (ix) a diabody.

Examples of different classes of bispecific antibodies include, but are not limited to: (i) IgG-like molecules with complementary CH3 domains to promote heterodimerization; (ii) dual-targeting recombinant IgG-like molecules in which each of the two sides of the molecule comprises a Fab fragment or portions of a Fab fragment of at least two different antibodies; (iii) IgG fusion molecules in which full-length IgG antibodies are fused to an additional Fab fragment or portions of a Fab fragment; (iv) Fc fusion molecules in which single-chain Fv molecules or stabilized diabodies are fused to heavy chain constant domains, Fc regions, or portions thereof; (v) Fab fusion molecules in which different Fab fragments are fused together, fused to heavy chain constant domains, Fc regions, or portions thereof; and (vi) ScFv and diabodies and heavy chain based antibodies (e.g. domain antibodies, Nanobodies®) in which different single chain Fv molecules or different diabodies or different heavy chain antibodies (e.g. domain antibodies, Nanobodies®) are fused to each other or to another protein or carrier molecule fused to heavy chain constant domains, Fc regions or portions thereof.

Examples of IgG-like molecules with complementary CH3 domains include, but are not limited to, Triomab® (Trion Pharma/Fresenius Biotech), Knobs-into-Holes (Genentech), CrossMAbs (Roche) and electrostatically selected antibodies (Amgen, Chugai, Oncomed), LUZ-Y (Genentech, Wranik et al., J. Biol. Chem. 2012, 287(52): 43331-9, doi: 10.1074/jbc.M112.397869. Epub 2012 Nov 1), DIG-body and PIG-body (Pharmabcine, WO2010134666, WO2014081202), chain exchange domain modified antibody (SEEDbody)(EMD Serono), Biclonics (Merus, WO2013157953), FcAAdp (Regeneron), bispecific IgG1 and IgG2 (Pfizer/Rinat), having an azimetric framework (Zymeworks/Merck,), mAb-Fv (Xencor), bivalent bispecific antibodies (Roche, WO2009080254) and DuoBody® molecules (Genmab).

Examples of recombinant dual-targeting IgG-like molecules include, but are not limited to, dual-targeting (DT)-Ig (GSK/Domantis, WO2009058383), Two-in-one antibody (Genentech, Bostrom, et al., 2009. Science 323, 1610-1614), cross-linked antibodies (Karmanos Cancer Center), mAb2 (F-Star), ZybodiesTM (Zyngenia, LaFleur et al., MAbs. 2013 Mar-Apr;5(2):208-18), common light chain-based approaches, κλ-body (NovImmune, WO2012023053), and CovX-body® (CovX/Pfizer, Doppalapudi, V.R., et al 2007). Bioorg. Med. Chem. Lett. 17.501-506).

Examples of IgG fusion molecules include, but are not limited to, dual variable domain (DVD) IgTMs (Abbott), dual head dual domain antibodies (Unilever; Sanofi Aventis), IgG-like bispecifics (ImClone/Eli Lilly, Lewis et al., Nat Biotechnol. 2014

- 7 048784

Feb;32(2):191-8), Ts2Ab (MedImmune/AZ, Dimasi et al., J Mol Biol. 2009 Oct 30;393(3):672-92) and BsAb (Zymogenetics, WO2010111625), HERCULES (Biogen Idec), scFv fusion product (Novartis), scFv fusion product (Changzhou Adam Biotech Inc) and TvAb (Roche).

Examples of Fc fusion molecules include, but are not limited to, ScFv/Fc fusion products (Academic Institution, Pearce et al Biochem Mol Biol Int. 1997 Sep;42(6):1179), SCORPION (Emergent BioSolutions/Trubion, Blankenship JW, et al., AACR 100th Annual meeting 2009 (Abstract #5465); Zymogenetics/BMS, WO2010111625), dual affinity antibody retargeting technology (FcDARTTM) (MacroGenics), and Dual(ScFv) ₂ -Fab (National Research Center for Antibody Medicine - China).

Examples of Fab fusion bispecific antibodies include, but are not limited to, F(ab) ₂ (Medarex/AMGEN), dual-action Fab or Bis-Fab (Genentech), Dock-and-Lock® (DNL) (ImmunoMedics), bivalent bispecific (Biotecnol), and Fab-Fv (UCB-Celltech).

Examples of antibodies that are based on ScFv, diabodies, and domain antibodies include, but are not limited to, bispecific T cell activator (BiTE®) (Micromet, Tandem Diabody (Tandab) (Affimed), dual affinity antibody retargeting technology (DARTTM) (MacroGenics), single chain antibody (Academic, Lawrence FEBS Lett. 1998 Apr 3;425(3):479-84), TCR-like antibodies (AIT, ReceptorLogics), human serum albumin ScFv fusion product (Merrimack, WO2010059315) and COMBODY molecules (Epigen Biotech, Zhu et al., Immunol Cell Biol. 2010 Aug;88(6):667-75), dual targeting nanobodies® (Ablynx, Hmila et al., FASEB J. 2010), dual-targeting heavy chain domain-only antibodies.

A binding fragment in the context of the present invention is a fragment, preferably a polypeptide, more preferably an antibody, which is capable of specifically binding to a target. A binding fragment may, for example, comprise: an intact immunoglobulin molecule, such as a monoclonal antibody. Alternatively, the binding fragment may comprise an antigen-binding functional fragment, including, but not limited to, Fab, F(ab'), F(ab') ₂ , Fv, dAb, Fd, complementarity determining region (CDR) fragment, single chain antibody (scFv), divalent single chain antibody, single chain phage antibody, bispecific double chain antibody, triabody, tetrabody, single domain antibody (nanobody®), (poly)peptide comprising at least an amino acid sequence sufficient for specific binding to its target, and artificial globulin fragments such as plastic antibodies (Hoshino et al. (2008) J Am Chem Soc 130(46): 15242). In a preferred embodiment, the binding fragment of the present invention is a single domain antibody. Preferably, the binding fragment of the present invention comprises three CDRs of the heavy chain.

The term "specifically binds" as used herein means binding of a binding moiety or binding molecule to a predetermined antigen or target (e.g. human CD1d), which binding typically occurs with an affinity corresponding to a KD value of about 10 ^-6 M or less, such as 10 ^-7 M or less, such as about 10 ^-8 M or less, such as about 10 ^-9 M or less, about 10 ^-10 M or less, or about 10 ^-11 M or even less, when determined using, for example, surface plasmon resonance (SPR) technology in a BIAcore 3000 instrument using the antigen as a ligand and the binding moiety or binding molecule as an analyte, and binds to the predetermined antigen with an affinity corresponding to a KD value of at least 10-fold lower, such as at least 100-fold lower, such as at least 1000-fold lower less, such as at least 10,000 times less, such as at least 100,000 times less, than its binding affinity to a non-specific antigen (e.g. BSA, casein) other than a predetermined antigen or a closely related antigen. The degree of affinity reduction depends on the KD of the binding moiety or binding molecule, so that when the KD of the binding moiety or binding molecule is very low (i.e. when the binding moiety or binding molecule is highly specific) the degree of reduction in affinity for the antigen compared to the affinity for the non-specific antigen may be at least 10,000-fold. The term KD (M) as used herein means the equilibrium constant for dissociation of a specific interaction between the antigen and the binding moiety or binding molecule.

As indicated, the above-described binding moiety is capable of competing for binding of human CD1d with the single domain antibody 1D12. In the context of the present invention, the terms "compete" or "is capable of competing" or "compete" mean any detectable, significant reduction in the propensity of a particular binding molecule (e.g., a CD1d antibody) to bind to a particular partner (e.g., CD1d) in the presence of another molecule (e.g., another CD1d antibody) that binds to that partner. Typically, "compete" means at least about 25% reduction, such as at least about 50%, such as at least about 75%, such as at least 90% reduction in binding between a CD1d binding molecule or moiety caused by the presence of another CD1d binding molecule or moiety.

- 8 048 784 molecule or fragment when determined, for example, by an ELISA assay or flow cytometry using sufficient quantities of two or more competing binding molecules or fragments. Additional methods for determining binding specificity by competitive inhibition can be found in, for example, Harlow et al., Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1988), Colligan et al., eds., Current Protocols in Immunology, Greene Publishing Assoc., and Wiley InterScience N. Y, (1992, 1993), and Muller, Meth. Enzymol. 92, 589601 (1983)). Preferably, the binding molecules of the present invention bind to at least a portion of the same epitope on CD1d as the antibody 1D12 or in the vicinity thereof, more preferably to the same epitope as 1D12 or to the same epitope that coincides with an epitope of 1D12. Methods for determining the epitope of a binding molecule, such as an antibody, are known in the art. By in the vicinity is meant that they bind to CD1d such that the binding molecule at least sterically hinders the binding of 1D12 to CD1d. A binding molecule of the present invention may comprise identical or different CDR sequences compared to 1D12. In a more preferred embodiment, a binding molecule of the present invention comprises sequences identical to CDR1 (SEQ ID NO 1), CDR2 (SEQ ID NO: 2) and CDR3 (SEQ ID NO: 3) of 1D12. In a most preferred embodiment, the binding molecule of the present invention comprises a sequence identical to the sequence of 1D12 (SEQ ID NO: 4) or comprises the sequence indicated in SEQ ID NO: 85.

As described above, in another main aspect, the invention provides a binding molecule comprising a first binding moiety that is capable of competing with 1D12 (the antibody set forth in SEQ ID NO: 4) for binding to a CD1d molecule and comprising a second binding moiety that is capable of specifically binding to Vγ9Vδ2-TCR. Specifically binding to Vγ9Vδ2-TCR means that the binding molecule binds Vγ9Vδ2-TCR, but does not exclude that the binding molecule binds to one of the individual subunits in the absence of the other subunit, i.e., the Vγ9 chain alone or the Vδ2 chain alone. For example, as shown in Table 2, antibody 5C8 is an antibody that binds to Vγ9Vδ2-TCR, but also binds to the Vδ2 chain when only the Vδ2 chain is expressed.

In a preferred embodiment of the invention, the binding molecule binds the Vδ2 chain of the Vγ9Vδ2 TCR. Moreover, preferably, the binding molecule is capable of activating Vγ9Vδ2 T cells independently of their natural ligand. The invention is based on the discovery that such a binding molecule, comprising a bispecific immunoglobulin complex, is capable of inhibiting Vδ1+ T cells and provides activation of (Vy9)V52+ T cells. Activation of Vγ9Vδ2 T cells is useful in the treatment of a variety of diseases. A Vδ2 TCR chain-specific immunoglobulin is preferred over a Vy9 TCR chain-specific immunoglobulin because Vy9 can bind to, for example, the V51 TCR chain, resulting in the recruitment and activation of potentially counterproductive γδ T cells.

The term first and second binding moiety does not refer to their orientation/position within the binding molecule, i.e. it does not matter in terms of N- or C-terminus. The term first and second only serves to correctly and consistently designate two different binding moieties in the claims and description of the invention. In one preferred embodiment, the first binding moiety and the second binding moiety are connected to each other by one or more amide bonds, preferably via a linker, more preferably comprising the amino acids GGGGS (SEQ ID NO: 83). Two non-limiting examples of binding molecules according to the invention comprising a first and a second binding enzyme connected by a peptide linker (GGGGS) are shown in Table 1 (SEQ ID NO: 84 and SEQ ID NO: 87). In another preferred embodiment, the binding molecule is a bispecific antibody, such as a full-length bispecific antibody. The term "full-length antibody" as used herein means an antibody that contains all of the constant and variable domains of the heavy and light chains corresponding to the domains normally found in a wild-type antibody of a given isotype.

By the phrase "capable of activating Vγ9Vδ2 T cells" in the context of the present invention it is meant that Vγ9Vδ2 T cells are activated in the presence of a binding molecule according to the invention, in particular in the presence of a target cell expressing CD1d. Preferably, the activation of Vγ9Vδ2 T cells can be measured on the basis of gene expression and/or (surface) expression of markers (e.g. activation markers such as CD25, CD69 or CD107a) and/or secretory protein profiles (e.g. cytokines or chemokines). In a preferred embodiment, the binding molecule is capable of inducing activation (e.g. increased expression of CD69 and/or CD25), resulting in degranulation (measured by increased expression of CD107a; Example 3) and production of cytokines (e.g. TNFa, IFNy) by Vγ9Vδ2 T cells. Preferably, the activation of Vγ9Vδ2 T cells occurs in vivo, in particular in the human body to which the binding molecule according to the invention has been administered, wherein the human body comprises

- 9 048 784 Vγ9Vδ2 T cells and preferably CD1d+ target cells. Preferably, a binding molecule of the present invention is capable of increasing the expression of CD107a on Vγ9Vδ2 T cells by at least 10%, more preferably by at least 20%, more preferably by at least 40%, most preferably by at least 90% when used in an assay as described in Example 3, wherein for example 10% means that 10% of the total number of cells are CD107a-positive. In another embodiment, the number of CD107a-positive cells is increased by 1.5-fold, such as 2-fold, such as 5-fold in the presence of a binding molecule of the invention.

Likewise, for iNKT cells, the phrase "able to activate" in the context of the present invention means that the iNKT cells behave differently in the presence of a binding molecule or a binding molecule for use according to the invention, in particular in the presence of a CD1d molecule, preferably in the presence of a CD1d molecule on the cell surface. Markers such as CD25 (example 1), CD69, CD107a (example 3), or cytokines/chemokines such as IFNy (example 1), TNFa, IL-2 are used to determine the activation of iNKT cells. Preferably, the activation of iNKT cells occurs in vivo, in particular in a human body to which a binding molecule according to the invention has been administered, wherein the human body comprises iNKT cells and preferably CD1d+ target cells. By CD1d+ target cells are meant CD1d+ cells that contribute to the pathogenicity of the disease and not normal CD1d-expressing cells. Preferably, the binding molecule of the present invention is capable of increasing the expression of CD107a on iNKT cells by at least 20%, more preferably by at least 30%, most preferably by at least 40% when used in an assay as described in Example 3, wherein, for example, 10% means that 10% of the total number of cells are CD107a-positive. In another embodiment, the number of CD107a-positive cells is increased by 1.5-fold, such as 2-fold, such as 5-fold in the presence of the binding molecule of the invention. Moreover, preferably, the binding molecule of the present invention is capable of increasing the expression of CD25 on iNKT cells by at least 10-fold, more preferably by at least 20-fold, most preferably by at least 30-fold compared to a vehicle control when measured by mean fluorescence intensity by flow cytometry using allophycocyanin (APC)-conjugated CD25 in FACS when used in the assay as described in Example 1. Preferably, the binding molecule of the present invention is capable of increasing the expression of IFNy by iNKT cells by at least 1.5-fold, such as at least 2-fold or at least 3-fold when used in the assay as described in Example 1.

The bispecific (or multispecific) binding molecule preferably induces a Th1-type response. In a preferred embodiment, a binding molecule comprising a first and a second binding moiety according to the invention is provided for use as a medicament, preferably for use in the treatment of a tumor. A tumor within the context of the present invention is understood to mean: an abnormal mass of tissue located in one place (e.g. solid tumors) or distributed throughout the human body (e.g. metastases). Furthermore, an abnormal mass of tissue is also understood to mean disseminated tumors (e.g. liquid hematological tumors). Tumors are also a classic sign of inflammation. However, such inflammatory (non-cancerous) tumors and other non-malignant tumors are not included in the definition herein. In a preferred embodiment, the tumor is cancer, especially a type that is potentially fatal. The treatment of tumors depends on the location and type of tumor. Benign tumors can sometimes simply be ignored or they can be reduced (partially removed) or completely removed by surgery. For malignant or some benign tumors, treatment options include chemotherapy, radiation, and surgery. Tumors of the hematopoietic and lymphoid tissues (called blood tumors) are also covered and include, inter alia, leukemias such as ALL, AML, CLL, small lymphocytic lymphoma (SLL), chronic myelogenous leukemia (CML), and acute monocytic leukemia (AMoL), lymphomas such as Hodgkin's lymphoma and non-Hodgkin's lymphoma, and myelomas.

In a preferred embodiment, there is provided a binding molecule comprising a first binding moiety according to the invention, or a binding molecule comprising a first and a second binding moiety, for use according to the invention, wherein the binding molecule is capable of reducing the activation of V51 T cells. Reducing the activation of Vδ1 T cells in this context means that the V51 T cell is no longer able to recognize its ligand on the CD1d molecule that is bound to the binding molecule described in the invention. Such reduced activation of Vδ1+ T cells can, for example, be determined by measuring the expression of the activation marker CD69 on Vδ1+ T cells. Lower levels of CD69 expression are observed when the activation of Vδ1+ T cells, such as Jurkat cells, is low, as described in Example 2. In particular, when using

- 10 048784 in the context of νδ1+ tumor cells, blocking means that the presence of the binding molecule according to the invention has a negative effect on the growth and/or viability of the tumor cells. Preferably, the expression of CD69 on Jurkat cells is increased by less than 5-fold, such as less than 2-fold, by the binding molecule according to the invention compared to the vehicle control when tested in the assay as described in Example 2.

Preferably, a binding molecule for use according to the invention is capable of activating iNKT cells. As indicated above, the inventors have shown that 1D12 is capable of activating iNKT cells regardless of the presence of an exogenous ligand for iNKT cells, such as α-galactosylceramide. The present invention is also based on the discovery that the recognition of CD1d by Vδ1+ T cells can be blocked by such binding molecules and that the presence of a second binding moiety in the binding molecule according to the invention ensures the activation of Vδ2+ T cells. Such a binding molecule with a triple effect, i.e. capable of reducing the activation of CD1d-restricted Vδ1+ T cells and activating iNKT cells and at the same time ensuring the activation of Vδ2+ T cells, creates a microenvironment with a shift towards a Th1 type antitumor response. Reduced Vδ1+ T cell activation, iNKT cell activation, and Vδ2+ T cell activation collectively form a tumor-aggressive microenvironment that enables effective tumor cell killing.

Thus, the binding molecule of the invention can be modified such that it not only reduces the activation of Vδ1+ T cells and activates iNKT cells, but also binds and activates Vδ2+ T cells by clustering Vγ9Vδ2-TCR on γδ-T cells. For this purpose, the binding molecule of the invention comprises a second binding moiety. This second binding moiety is capable of binding to Vγ9Vδ2-TCR, preferably to the Vδ2 chain. Some such antibodies that bind to the Vδ2 chain or the V γ9 chain of TCR have been described in WO2015156673 and are shown in Table 1 and Table 2. In another aspect, the invention provides a method of treating a subject in need thereof, comprising administering to the subject a first binding molecule of the invention comprising a first and a second binding moiety, wherein the first moiety binds to CD1d and the second binding moiety binds to Vγ9Vδ2 TCR, in combination with a second binding molecule as described above, comprising a binding moiety that binds to CD1d but does not include a binding moiety that binds to Vγ9Vδ2 TCR. Such combination therapy may be particularly useful if the first and second binding molecule are effective at very different concentrations. For example, an excess of CD1d antibody, which is blocked at a high concentration, may be combined with a CD1d/Vγ9Vδ2 antibody, which activates γδ T cells at a low concentration.

Table 1

Designation of VHH (CDR), TCR chains, and sequences of different VHH included in a binding molecule according to the invention

SEQ ID. code Description Subsequence 1 1D12 CDR1 DNVMG 2 1D12 CDR2 TIRTGGSTNYADSVKG 3 1D12 CDR3 TIPVPSTPYDY 4 1D12 QVQLVESGGGLVQAGGSLRLSCAASGSMFDNVM GWYRQAPGKQRELVATIRTGGSTNYADSVKGRFT ISRDNAKNTVYLQMNSLKPEDTAVYYCRHTIPVP STPYDYWGQGTQVTVSS 5 5E3 CDR1 SYAMG 6 5E3 CDR2 AISWSGGTTYYADSVKG

- 11 048784

7 5E3 CDR3 SLDCSGPGCHTAEYDY 8 6H1 CDR1 EYAMG 9 6H1 CDR2 AISWTGSKTYYADSVKG 10 6H1 CDR3 SSDCSGPGCHTEEYDY 11 5G3 CDR1 SYAMG 12 5G3 CDR2 AVSWSGGSTYYADSVKG 13 5G3 CDR3 SQDCSGPGCYTNEYDS 14 5C1 CDR1 NYAMA 15 5C1 CDR2 AVSWSGGRTYYADSVKG 16 5C1 CDR3 SLSCSGPGCSLEEYDY 17 5D3 CDR1 NYAMG 18 5D3 CDR2 VISWSGGSTYYADSVKG 19 5D3 CDR3 QFSGASTVVAGTALDYDY 20 6E3 CDR1 NYGMG 21 6E3 CDR2 GISWSGGSTDYADSVKG 22 6E3 CDR3 VFSGAETAYYPSDDYDY 23 6H4 CDR1 NYGMG 24 6H4 CDR2 GISWSGGSTDYADSVKG 25 6H4 CDR3 VFSGAETAYYPSDDYDY 26 6C1 CDR1 NYGMG 27 6C1 CDR2 GISWSGGSTDYADSVKG 28 6C1 CDR3 VFSGAETAYYPSDDYDY 29 6H3 CDR1 NYGMG 30 6H3 CDR2 GITWSGGSTHYADLVKG 31 6H3 CDR3 VFSGAETAYYPSTEYDY 32 6G3 CDR1 NYGMG 33 6G3 CDR2 GISWSGGSTYYADSVKG 34 6G3 CDR3 VFSGAETAQYPSYDYDY

- 12 048784

35 5C8 CDR1 NYAMG 36 5C8 CDR2 AISWSGGSTSYADSVKG 37 5C8 CDR3 QFSGADYGFGRLGIRGYEYDY 38 5F5 CDR1 NYAMG 39 5F5 CDR2 AISWSGGSTYYADSVKG 40 5F5 CDR3 MFSGSESQLVVVITNLYEYDY 41 6A1 CDR1 NYAMG 42 6A1 CDR2 TISWSGGSTYYADSVKG 43 6A1 CDR3 AFSGSDYANTKKEVEYDY 44 6E4 CDR1 DYCIA 45 6E4 CDR2 CITTSDGSTYYADSVKG 46 6E4 CDR3 Y FGYGCYGGAQDYRAMDY 47 5C7 CDR1 RYTMG 48 5C7 CDR2 AISWSGGRTNFAGSVKG 49 5C7 CDR3 DWLPVPGRESYDY 50 5D7 CDR1 NYAMG 51 5D7 CDR2 AISWSGGMTDHADSVKG 52 5D7 CDR3 AFAGDIPYGSSWYGDPTTYDY 53 5B11 CDR1 TFSMA 54 5B11 CDR2 AINWSGGSTRYADSVSD 55 5B11 CDR3 RRGGIYYSTQNDYDY 56 6C4 CDR1 DYRMG 57 6C4 CDR2 TISWSGGLTYYADSVKG 58 6C4 CDR3 GGGYAGGTYYHPEE 59 5E3 VHH EVQLVESGGGLVQAGGSLRLSCTASGRTFSSYAM GWFRQAPGKEREFVAAISWSGGTTYYADSVKGRF TISRDNAKNTVSLQMNSLKPEDTAVY FCAASLDC SGPGCHTAEYDYWGQGTQVTVSS 60 6H1 VHH EVQLVESGGGLVQAGGSLRLSCAATGRTFSEYAM

- 13 048784

GWFRQAPGKEREFAAAISWTGSKTYYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAASSDC SGPGCHTEEYDYWGQGTQVTVSS 61 5G3 VHH EVQLVESGGGLVQAGGSLRLSCAASGRTFSSYAM GWFRQAPGKEREFVAAVSWSGGSTYYADSVKGRF TISRDNARNTVYLQMNSLNPEDTAVYYCAASQDC SGPGCYTNEYDSWGQGTQVTVSS 62 5C1 VHH EVQLVESGGGLVQPGGSLRLSCAASGSIFSNYAM AWFRQAPEKERDFLAAVSWSGGRTYYADSVKGRF TISRDNAKNTVNLQMNSLKPEDTAVYYCAASLSC SGPGCSLEEYDYWGQGTQVTVSS 63 5D3 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFSNYAM GWFRQAPGKEREFVTVISWSGGSTYYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAQFSG ASTVVAGTALDYDYWGQGTRVTVSS 64 6E3 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFSNYGM GWFRQAPGKKREFVAGISWSGGSTDYADSVKGRL TISRDNAKNTVYLQMNSLKPEDTAVYYCAAVFSG AETAYYPSDDYDYWGQGTQVTVSS 65 6H4 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFSNYGM GWFRQAPGKKREFVAGISWSGGSTDYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAVFSG AETAYYPSDDYDYWGQGTQVTVSS 66 6C1 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFSNYGM GWFRQAPGKKRESVAGISWSGGSTDYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAVFSG AETAYYPSDDYDYWGQGTQVTVSS 67 6H3 VHH EVQLVESGGGLVQAGGSLRLSCAVSGRPFSNYGM GWFRQAPGKEREFVAGITWSGGSTHYADLVKGRF

- 14 048784

TISRDNAKNTVHLQMNSLKPEDTAVYYCAAVFSG AETAYYPSTEYDYWGQGTQVTVSS 68 6G3 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFNNYGM GWFRQAPGKEREFVAGISWSGGSTYYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAVFSG AETAQYPSYDYDYWGQGTQVTVSS 69 5C8 VHH EVQLVESGGGLVQAGGSLRLSCAASGRPFSNYAM GWFRQAPGKEREFVAAISWSGGSTSYADSVKGRF TISRDNAKNTVYLQMNSPKPEDTAIYYCAAQFSG ADYGFGRLGIRGYEYDYWGQGTQVTVSS 70 5F5 VHH EVQLVESGGGLVQAGGSLRLSCAASGRTFSNYAM GWFRQAPGKEREFVAAISWSGGSTYYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAMFSG SESQLVVVITNLYEYDYWGQGTQVTVSS 71 6A1 VHH EVQLVESGGGLVQAGGSLRLSCAASGRTFSNYAM GWFRQAPGKEREFVATISWSGGSTYYADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAAFSG SDYANTKKEVEYDYWGQGTQVTVSS 72 6E4 VHH EVQLVESGGGLVQAGGSLRLSCAASGFTFDDYCI AWFRQAPGKEREPVSCITTSDGSTYYADSVKGRF TIS S DNAKNTVYLQMNRL KP E DTAVY YCAAY FGY GCYGGAQDYRAMDYWGKGTLVTVSS 73 5C7 VHH EVQLVESGGGLVQAGDSLRLSCAASGRTFSRYTM GWFRQAPGKEREFVAAIWSSGRTNFAGSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAADWLP VPGRESYDYWGQGTQVTVSS 74 5D7 VHH EVQLVESGGGLVQAGGSLRLSCIASGRTFFSNYAM GWFRQAPGKEREFVAAIWSSGMTDHADSVKGRF TISRDNAKNTVYLQMNSLKPEDTAVYYCAAAFAG

- 15 048784

DIPYGSSWYGDPTTYDYWGQGTQVTVSS 75 5B11 VHH EVQLVESGGGLVQAGGSLRLSCAASGRTSSTFSM AWFRQAPRKEREFVAAINWSGGSTRYADSVSDRF AISRDNAKNTVYLQMNNLKPEDTAVYYCAARRGG IYYSTQNDYDYWGQGTQVTVSS 76 6C4 VHH EVQLVESGGGLVQAGGSLRLSCAVSVRTFSDYRM GWFRQAPGKEREFVSTISSWSGGLTYYADSVKGRF TISRDNSKNTLYLQMNSLKPEDTAVYYCAAGGGY AGGTYYHPEEWGQGTQVTVSS 77 Vy9 human chain TCR MLSLLHASTLAVLGALCVYGAGHLEQPQISSTKT LSKTARLECVVSGITISATSVYWYRERPGEVIQF LVSISYDGTVRKESGIPPSGKFEVDRIPETSTSTL TIHNVEKQDIATYYCALWEAQQELGKKIKVFGPG TKLIITDKQLDADVSPKPTIFLPSIAETKLQKAG TYLCLLEKFFPDVIKIHWEEKKSNTILGSQEGNT MKTNDTYMKFSWLTVPEKSLDKEHRCIVRHENNK NGVDQEIIFPPIKTDVITMDPKDNCSKDANDTLL LQLTNTSAYYMYLLLLLKSVVYFAIITCCLLRRT AFCCNGEKS 78 V52 human chain TCR MQRISSLIHLSLFWAGVMSAIELVPEHQTVPVSI GVPAT LRC SMKGEAIGNYYINWΥ RKT QGNTMT FI YREKDIYGPGFKDNFQGDIDIAKNLAVLKILAPS ERDEGSYYCACDTLGMGGEYTDKLIFGKGTRVTV EPRSQPHTKPSVFVMKNGTNVACLVKEFYPKDIR INLVSSKKITEFDPAIVISPSGKYNAVKLGKYED SNSVTCSVQHDNKTVHSTDFEVKTDSTDHVKPKE TENTKQPSKSCHKPKAIVHTEKVNMMSLTVLGLR ML FAKTVAVN FLLTAKL F FL

- 16 048784

79 1D22 CDR1 NAMG 80 1D22 CDR2 VISSSGSTNYADSVKG 81 1D22 CDR3 HVAGFDEYNY 82 1D22 VHH QVQLVESGGGLVQAGGSLRLSCAASGSIFSINAM GWYRQAPGKQRDFLAVISSSGSTNYADSVKGRFT IS RDNAKNTAY LQMN SLKVE DTAVYY CAAHVAG F DEYNYWGQGTQVTVSS 83 GS-linker Linker GGGGS 84 1D12- 5С8 Bispecific binding molecule QVQLVESGGGLVQAGGSLRLSCAASGSMFDNVM GWYRQAPGKQRELVATIRTGGSTNYADSVKGRFT ISRDNAKNTVYLQMNSLKPEDTAVYYCRHTIPVP STPYDYWGQGTQVTVSSGGGGSEVQLVESGGGLV QAGGSLRLSCAASGRPFSNYAMGWFRQAPGKERE FVAAISWSGGSTSYADSVKGRFTISRDNAKNTVY LQMNSPKPEDTAIYYCAAQFSGADYGFGRLGIRG YEYDYWGQGTQVTVSS 85 1D12 var VHH EVQLVESGGGLVQAGGSLRLSCAASGSMFDNVM GWYRQAPGKQRELVATIRTGGSTNYADSVKGRFT ISRDNAKNTVYLQMNSLKPEDTAVYYCRHTIPVP STPYDYWGQGTQVTVSS 86 5С8 var 1 VHH EVQLLESGGGSVQPGGSLRLSCAASGRPFSNYAM SWFRQAPGKEREFVSAISWSGGSTSYADSVKGRF TISRDNSKNTLYLQMNSLRAEDTAVYYCAAQFSG ADYGFGRLGIRGYEYDYWGQGTQVTVSS 87 1D12-5C8 var Bispecific binding molecule EVQLVESGGGLVQAGGSLRLSCAASGSMFDNVM GWYRQAPGKQRELVATIRTGGSTNYADSVKGRFT ISRDNAKNTVYLQMNSLKPEDTAVYYCRHTIPVP STPYDYWGQGTQVTVSSGGGGSEVQLLESGGGSV QPGGSLRLSCAASGRPFSNYAMSWFRQAPGKERE FVSAISWSGGSTSYADSVKGRFTISRDNSKNTLY LQMNSLRAEDTAVYYCAAQFSGADYGFGRLGIRG YEYDYWGQGTQVTVSS 88 5С8 var 2 VHH EVQLLESGGGLVQPGGSLRLSCAASGRPFSNYAM SWFRQAPGKEREFVSAISWSGGSTSYADSVKGRF TISRDNSKNTLYLQMNSLRAEDTAVYYCAAQFSG ADYGFGRLGIRGYEYDYWGQGTLVTVSS

From WO2015156673: binding of VHH to γδ T cells expressing the Vy9 or νδ2 chain (linked to a non-Y52 or hc-Vy9 chain, respectively), expressing the full Vγ9Vδ2 TCR or not expressing any of the Vγ9Vδ2 TCR chains.

- 17 048784

Table 2

SEQ ID Ref V62 + Vy9 + Vy9V62 + Vy9V62 - 59 5E3 - ++ ++ - 60 6N1 - ++ ++ - 61 5G3 - ++ ++ - 62 5C1 +/- ++ ++ - 63 5D3 ++ - ++ - 64 6E3 ++ - ++ - 65 6N4 ++ - ++ - 66 6C1 ++ - ++ - 67 6NZ ++ +/- ++ - 68 6G3 ++ - ++ - 69 5C8 ++ - ++ - 70 5F5 ++ - ++ - 71 6A1 ++ - ++ - 72 6E4 ++ - ++ - 73 5C7 +/- - +/- - 74 5D7 ++ - ++ - 75 5B11 - - + - 76 6C4 +/- ++ ++ -

- Indicates mean fluorescence index (MFI) below 1.5, +/- indicates MFI from 1.5 to 4.5, + indicates MFI from 4.6 to 20, and ++ indicates MFI above 20.

Thus, there is provided a binding molecule according to the invention or a binding molecule for use according to the invention, wherein the second binding moiety is capable of competing for binding of the Vγ9Vδ2 TCR with a single-domain antibody of 5E3 (SEQ ID NO: 59), 6H1 (SEQ ID NO: 60), 5G3 (SEQ ID NO: 61), 5C1 (SEQ ID NO: 62), 5D3 (SEQ ID NO: 63), 6E3 (SEQ ID NO: 64), 6H4 (SEQ ID NO: 65), 6C1 (SEQ ID NO: 66), 6H3 (SEQ ID NO: 67), 6G3 (SEQ ID NO: 68), 5C8 (SEQ ID NO: 69), 5F5 (SEQ ID NO: 70), 6A1 (SEQ ID NO: 71), 6E4 (SEQ ID NO: 72), 5C7 (SEQ ID NO: 73), 5D7 (SEQ ID NO: 74), 5B11 (SEQ ID NO: 75), or 6C4 (SEQ ID NO: 76), preferably with a single domain antibody of 5E3 (SEQ ID NO: 59), 6H1 (SEQ ID NO: 60), 5G3 (SEQ ID NO: 61), 5C1 (SEQ ID NO: 62), 5D3 (SEQ ID NO: 63), 6E3 (SEQ ID NO: 64), 6H4 (SEQ ID NO: 65), 6C1 (SEQ ID NO: 66), 6H3 (SEQ ID NO: 67), 6G3 (SEQ ID NO: 68), 5C8 (SEQ ID NO: 69), 5F5 (SEQ ID NO: 70), 6A1 (SEQ ID NO: 69), 6H2 (SEQ ID NO: 71), 6H3 (SEQ ID NO: 72), 5D7 (SEQ ID NO: 74), 5B11 (SEQ ID NO: 75), or 6C4 (SEQ ID NO: 76), preferably with a single domain antibody of 5E3 (SEQ ID NO: 59), 6H1 (SEQ ID NO: 60), 5G3 (SEQ ID NO: 61), 5C1 (SEQ ID NO: 62), 5D3 (SEQ ID NO: 63), 6E3 (SEQ ID NO: 64), 6H4 (SEQ ID NO: 65), 6C1 (SEQ ID NO: 66), 6H3 (SEQ ID NO: 67), 6G3 (SEQ ID NO: 68), 5C8 (SEQ ID NO: 69), 5F5 (SEQ ID NO: 70), 6A1 (SEQ ID NO: ID NO: 71), or 6E4 (SEQ ID NO: 72). Preferably, the second binding moiety binds to the same or a partially matching, more preferably the same, epitope sequence that is recognized by the binding moiety of 5E3, 6H1, 5G3, 5C1, 5D3, 6E3, 6H4, 6C1, 6H3, 6G3, 5C8, 5F5, 6A1, 6E4, 5C7, 5D7, 5B11 or 6C4, preferably which is recognized by the binding moiety of 5E3, 6H1, 5G3, 5C1, 5D3, 6E3, 6H4, 6C1, 6H3, 6G3, 5C8, 5F5, 6A1, or 6E4.

In some embodiments, the binding molecule of the invention or the binding molecule for use according to the invention, the first binding moiety or the second binding moiety, or both, is a single-domain antibody. Single-domain antibodies (sdAbs, also called Nanobody® or VHH) are well known to those skilled in the art.

Single-domain antibodies include one CDR1, one CDR2, and one CDR3. Examples of single-domain antibodies are variable fragments of heavy-chain-only antibodies, i.e., antibodies that do not naturally include light chains, single-domain antibodies derived from traditional antibodies, and modified antibodies. Single-domain antibodies can be derived from

- 18 048784 any species, including mouse, human, camel, llama, shark, goat, rabbit and cow. For example, naturally occurring VHH molecules can be derived from antibodies grown in camelids, such as camels, dromedaries, alpacas and guanacos.

Like a whole antibody, a single-domain antibody is capable of selectively binding to a specific antigen. Single-domain antibodies may contain only the variable domain of an immunoglobulin chain, i.e., CDR1, CDR2, and CDR3 and the framework regions. With a molecular weight of only about 12-15 kDa, single-domain antibodies are much smaller than conventional antibodies (150-160 kDa), which are composed of two heavy chains and two light chains, or even Fab fragments (53 kDa), which consist of one light chain and part of a heavy chain. The single-domain antibody format has the advantage of less steric hindrance when binding to its target.

In a preferred embodiment, there is provided a binding molecule according to the invention or a binding molecule for use according to the invention, wherein the first binding moiety comprises the CDR1 sequence GSMFSDNVMG (SEQ ID NO: 1), the CDR2 sequence TIRTGGSTNYADSVKG (SEQ ID NO: 2), and/or the CDR3 sequence TIPVPSTPYDY (SEQ ID NO: 3), or any of these sequences, wherein independently any of the amino acids, preferably no more than 4 amino acids, more preferably no more than 3, more preferably no more than 2, more preferably no more than 1 have been substituted, preferably conservatively substituted, according to Table 3.

In another preferred embodiment, there is provided a binding molecule according to the invention or a binding molecule for use according to the invention, wherein the first binding moiety comprises the sequence indicated in SEQ ID NO: 4 or the sequence indicated in SEQ ID NO: 85.

Table 3

Conservative amino acid substitutions

Remainder Conservative substitutions Remainder Conservative substitutions Ala Ser Leu lie; Vai Arg Lys Lys Arg; Gin Asn Gin; His Met Leu; lie Asp Glu Phe Met; Leu; Tyr Gin Asn Ser Thr, Gly Cys Ser Thr Sir; Vai Glu Asp Trp Tyr Gly Pro Tyr Trp; Phe His Asn; Gin Vai lie; Leu lie Leu, Vai

In a preferred embodiment, there is provided a binding molecule according to the invention or a binding molecule for use according to the invention, wherein the second binding moiety comprises a CDR 1, CDR2 and CDR3 sequence in the combinations shown in Table 1 for each VHH. More particularly, the second binding moiety comprises a CDR 1 sequence according to SEQ ID NO: 5, a CDR 2 sequence according to SEQ ID NO: 6 and/or a CDR 3 sequence according to SEQ ID NO: 7; or a CDR 1 sequence according to SEQ ID NO: 8, a CDR 2 sequence according to SEQ ID NO: 9 and/or a CDR 3 sequence according to SEQ ID NO: 10; or a CDR 1 sequence according to SEQ ID NO: 11, a CDR 2 sequence according to SEQ ID NO: 12 and/or a CDR 3 sequence according to SEQ ID NO: 13; or CDR 1 sequence according to SEQ ID NO: 14, CDR 2 sequence according to SEQ ID NO: 15 and/or CDR 3 sequence according to SEQ ID NO: 16; or CDR 1 sequence according to SEQ ID NO: 17, CDR 2 sequence according to SEQ ID NO: 18 and/or CDR 3 sequence according to SEQ ID NO: 19; or CDR 1 sequence according to SEQ ID NO: 20, CDR 2 sequence according to SEQ ID NO: 21 and/or CDR 3 sequence according to SEQ ID NO: 22; or CDR 1 sequence according to SEQ ID NO: 23, CDR 2 sequence according to SEQ ID NO: 24 and/or CDR 3 sequence according to SEQ ID NO: 25; or CDR 1 sequence according to SEQ ID NO: 26, CDR 2 sequence according to SEQ ID NO: 27 and/or CDR 3 sequence according to SEQ ID NO: 28; or CDR 1 sequence according to SEQ ID NO: 29, CDR 2 sequence according to SEQ ID NO: 30 and/or CDR 3 sequence according to SEQ ID NO: 31; or CDR 1 sequence according to SEQ ID NO: 32, CDR 2 sequence according to SEQ ID NO: 33 and/or CDR 3 sequence according to SEQ ID NO: 34; or CDR 1 sequence according to SEQ ID NO: 35, CDR 2 sequence according to

- 19 048784

SEQ ID NO: 36 and/or CDR 3 sequence according to SEQ ID NO: 37; or CDR 1 sequence according to SEQ ID NO: 38, CDR 2 sequence according to SEQ ID NO: 39 and/or CDR 3 sequence according to SEQ ID NO: 40; or CDR 1 sequence according to SEQ ID NO: 41, CDR 2 sequence according to SEQ ID NO: 42 and/or CDR 3 sequence according to SEQ ID NO: 43; or CDR 1 sequence according to SEQ ID NO: 44, CDR 2 sequence according to SEQ ID NO: 45 and/or CDR 3 sequence according to SEQ ID NO: 46; or CDR 1 sequence according to SEQ ID NO: 47, CDR 2 sequence according to SEQ ID NO: 48 and/or CDR 3 sequence according to SEQ ID NO: 49; or the CDR 1 sequence according to SEQ ID NO: 50, the CDR 2 sequence according to SEQ ID NO: 51 and/or the CDR 3 sequence according to SEQ ID NO: 52; or the CDR 1 sequence according to SEQ ID NO: 53, the CDR 2 sequence according to SEQ ID NO: 54 and/or the CDR 3 sequence according to SEQ ID NO: 55; or the CDR 1 sequence according to SEQ ID NO: 56, the CDR 2 sequence according to SEQ ID NO: 57 and/or the CDR 3 sequence according to SEQ ID NO: 58 or any of these sequences, wherein, independently, any amino acids, preferably no more than 4 amino acids, more preferably no more than 3, more preferably no more than 2, most preferably no more than 1, are conservatively substituted according to Table 3.

More preferably, the second binding fragment comprises a CDR 1 sequence according to SEQ ID NO: 5, a CDR 2 sequence according to SEQ ID NO: 6 and/or a CDR 3 sequence according to SEQ ID NO: 7; or a CDR 1 sequence according to SEQ ID NO: 8, a CDR 2 sequence according to SEQ ID NO: 9 and/or a CDR 3 sequence according to SEQ ID NO: 10; or a CDR 1 sequence according to SEQ ID NO: 11, a CDR 2 sequence according to SEQ ID NO: 12 and/or a CDR 3 sequence according to SEQ ID NO: 13; or a CDR 1 sequence according to SEQ ID NO: 14, a CDR 2 sequence according to SEQ ID NO: 15 and/or a CDR 3 sequence according to SEQ ID NO: 16; or CDR 1 sequence according to SEQ ID NO: 17, CDR 2 sequence according to SEQ ID NO: 18 and/or CDR 3 sequence according to SEQ ID NO: 19; or CDR 1 sequence according to SEQ ID NO: 20, CDR 2 sequence according to SEQ ID NO: 21 and/or CDR 3 sequence according to SEQ ID NO: 22; or CDR 1 sequence according to SEQ ID NO: 23, CDR 2 sequence according to SEQ ID NO: 24 and/or CDR 3 sequence according to SEQ ID NO: 25; or CDR 1 sequence according to SEQ ID NO: 26, CDR 2 sequence according to SEQ ID NO: 27 and/or CDR 3 sequence according to SEQ ID NO: 28; or CDR 1 sequence according to SEQ ID NO: 29, CDR 2 sequence according to SEQ ID NO: 30 and/or CDR 3 sequence according to SEQ ID NO: 31; or CDR 1 sequence according to SEQ ID NO: 32, CDR 2 sequence according to SEQ ID NO: 33 and/or CDR 3 sequence according to SEQ ID NO: 34; or CDR 1 sequence according to SEQ ID NO: 35, CDR 2 sequence according to SEQ ID NO: 36 and/or CDR 3 sequence according to SEQ ID NO: 37; or CDR 1 sequence according to SEQ ID NO: 38, CDR 2 sequence according to SEQ ID NO: 39 and/or CDR 3 sequence according to SEQ ID NO: 40; or the CDR 1 sequence according to SEQ ID NO: 41, the CDR 2 sequence according to SEQ ID NO: 42 and/or the CDR 3 sequence according to SEQ ID NO: 43; or the CDR 1 sequence according to SEQ ID NO: 44, the CDR 2 sequence according to SEQ ID NO: 45 and/or the CDR 3 sequence according to SEQ ID NO: 46, or any of these sequences, wherein, independently, any amino acids, preferably no more than 4 amino acids, more preferably no more than 3, more preferably no more than 2, most preferably no more than 1, are conservatively substituted according to Table 3.

In a preferred embodiment of the invention there is provided a binding molecule for use according to the invention, comprising a binding moiety capable of specifically binding to CD1d, wherein the first binding moiety comprises a CDR1 sequence according to SEQ ID NO: 1, a CDR2 sequence according to SEQ ID NO: 2 and a CDR3 sequence according to SEQ ID NO: 3.

In a preferred embodiment of the invention, there is provided a binding molecule comprising a first binding moiety capable of specifically binding to CD1d and comprising a second binding moiety that is capable of specifically binding to the Vγ9Vδ2 TCR of the invention, wherein:

the first binding moiety comprises a CDR1 sequence according to SEQ ID NO: 1, a CDR2 sequence according to SEQ ID NO: 2 and a CDR3 sequence according to SEQ ID NO: 3, and the second binding moiety comprises a CDR 1 sequence according to SEQ ID NO: 5, a CDR 2 sequence according to SEQ ID NO: 6 and a CDR 3 sequence according to SEQ ID NO: 7; or a CDR 1 sequence according to SEQ ID NO: 8, a CDR 2 sequence according to SEQ ID NO: 9 and a CDR 3 sequence according to SEQ ID NO: 10; or a CDR 1 sequence according to SEQ ID NO: 11, a CDR 2 sequence according to SEQ ID NO: 12 and a CDR 3 sequence according to SEQ ID NO: 13; or a CDR 1 sequence according to SEQ ID NO: 14, a CDR 2 sequence according to SEQ ID NO: 15 and a CDR 3 sequence according to SEQ ID NO: 16; or CDR 1 sequence according to SEQ ID NO: 17, CDR 2 sequence according to SEQ ID NO: 18 and CDR 3 sequence

- 20 048784 the sequence according to SEQ ID NO: 19; or the CDR 1 sequence according to SEQ ID NO: 20, the CDR 2 sequence according to SEQ ID NO: 21 and the CDR 3 sequence according to SEQ ID NO: 22; or the CDR 1 sequence according to SEQ ID NO: 23, the CDR 2 sequence according to SEQ ID NO: 24 and the CDR 3 sequence according to SEQ ID NO: 25; or the CDR 1 sequence according to SEQ ID NO: 26, the CDR 2 sequence according to SEQ ID NO: 27 and the CDR 3 sequence according to SEQ ID NO: 28; or the CDR 1 sequence according to SEQ ID NO: 29, the CDR 2 sequence according to SEQ ID NO: 30 and the CDR 3 sequence according to SEQ ID NO: 31; or CDR 1 sequence according to SEQ ID NO: 32, CDR 2 sequence according to SEQ ID NO: 33 and CDR 3 sequence according to SEQ ID NO: 34; or CDR 1 sequence according to SEQ ID NO: 35, CDR 2 sequence according to SEQ ID NO: 36 and CDR 3 sequence according to SEQ ID NO: 37; or CDR 1 sequence according to SEQ ID NO: 38, CDR 2 sequence according to SEQ ID NO: 39 and CDR 3 sequence according to SEQ ID NO: 40; or CDR 1 sequence according to SEQ ID NO: 41, CDR 2 sequence according to SEQ ID NO: 42 and CDR 3 sequence according to SEQ ID NO: 43; or CDR 1 sequence according to SEQ ID NO: 44, CDR 2 sequence according to SEQ ID NO: 45 and CDR 3 sequence according to SEQ ID NO: 46; or CDR 1 sequence according to SEQ ID NO: 47, CDR 2 sequence according to SEQ ID NO: 48 and CDR 3 sequence according to SEQ ID NO: 49; or CDR 1 sequence according to SEQ ID NO: 50, CDR 2 sequence according to SEQ ID NO: 51 and CDR 3 sequence according to SEQ ID NO: 52; or CDR 1 sequence according to SEQ ID NO: 53, CDR 2 sequence according to SEQ ID NO: 54 and CDR 3 sequence according to SEQ ID NO: 55; or CDR 1 sequence according to SEQ ID NO: 56, CDR 2 sequence according to SEQ ID NO: 57 and CDR 3 sequence according to SEQ ID NO: 58.

In one embodiment of the invention, there is provided a binding molecule comprising a binding moiety capable of specifically binding to CD1d for use in the treatment of diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells, preferably for use in the treatment of peripheral CD1d-restricted Vδ1+ T cell lymphoma, wherein the first binding moiety comprises a CDR1 sequence according to SEQ ID NO: 1, a CDR2 sequence according to SEQ ID NO: 2 and a CDR3 sequence according to SEQ ID NO: 3.

In one embodiment, the invention relates to a binding molecule comprising a first binding moiety capable of specifically binding to CD1d and comprising a second binding moiety that is capable of specifically binding to Vγ9Vδ2 TCR, for use in the treatment of diseases caused, maintained and/or promoted by the activation of CD1d-restricted Vδ1+ T cells, preferably for use in the treatment of peripheral CD1d-restricted Vδ1+ T cell lymphoma, wherein the first binding moiety comprises a CDR1 sequence according to SEQ ID NO: 1, a CDR2 sequence according to SEQ ID NO: 2 and a CDR3 sequence according to SEQ ID NO: 3, and the second binding moiety comprises a CDR 1 sequence according to SEQ ID NO: 5, a CDR 2 sequence according to SEQ ID NO: 6 and a CDR 3 sequence according to SEQ ID NO: 7; or CDR 1 sequence according to SEQ ID NO: 8, CDR 2 sequence according to SEQ ID NO: 9 and CDR 3 sequence according to SEQ ID NO: 10; or CDR 1 sequence according to SEQ ID NO: 11, CDR 2 sequence according to SEQ ID NO: 12 and CDR 3 sequence according to SEQ ID NO: 13; or CDR 1 sequence according to SEQ ID NO: 14, CDR 2 sequence according to SEQ ID NO: 15 and CDR 3 sequence according to SEQ ID NO: 16; or CDR 1 sequence according to SEQ ID NO: 17, CDR 2 sequence according to SEQ ID NO: 18 and CDR 3 sequence according to SEQ ID NO: 19; or CDR 1 sequence according to SEQ ID NO: 20, CDR 2 sequence according to SEQ ID NO: 21 and CDR 3 sequence according to SEQ ID NO: 22; or CDR 1 sequence according to SEQ ID NO: 23, CDR 2 sequence according to SEQ ID NO: 24 and CDR 3 sequence according to SEQ ID NO: 25; or CDR 1 sequence according to SEQ ID NO: 26, CDR 2 sequence according to SEQ ID NO: 27 and CDR 3 sequence according to SEQ ID NO: 28; or CDR 1 sequence according to SEQ ID NO: 29, CDR 2 sequence according to SEQ ID NO: 30 and CDR 3 sequence according to SEQ ID NO: 31; or CDR 1 sequence according to SEQ ID NO: 32, CDR 2 sequence according to SEQ ID NO: 33 and CDR 3 sequence according to SEQ ID NO: 34; or CDR 1 sequence according to SEQ ID NO: 35, CDR 2 sequence according to SEQ ID NO: 36 and CDR 3 sequence according to SEQ ID NO: 37; or CDR 1 sequence according to SEQ ID NO: 38, CDR 2 sequence according to SEQ ID NO: 39 and CDR 3 sequence according to SEQ ID NO: 40; or CDR 1 sequence according to SEQ ID NO: 41, CDR 2 sequence according to SEQ ID NO: 42 and CDR 3 sequence according to SEQ ID NO: 43; or CDR 1 sequence according to SEQ ID NO: 44, CDR 2 sequence according to SEQ ID NO: 45 and CDR 3 sequence according to SEQ ID NO: 46; or CDR 1 sequence according to SEQ ID NO: 47, CDR 2 sequence according to SEQ ID NO: 48 and CDR 3 sequence according to SEQ ID NO: 49; or CDR 1 after

- 21 048784 sequence according to SEQ ID NO: 50, CDR 2 sequence according to SEQ ID NO: 51 and CDR 3 sequence according to SEQ ID NO: 52; or CDR 1 sequence according to SEQ ID NO: 53, CDR 2 sequence according to SEQ ID NO: 54 and CDR 3 sequence according to SEQ ID NO: 55; or CDR 1 sequence according to SEQ ID NO: 56, CDR 2 sequence according to SEQ ID NO: 57 and CDR 3 sequence according to SEQ ID NO: 58.

In a more preferred embodiment of the invention, the second binding moiety comprises a sequence selected from any of SEQ ID NOs: 59-76, 86 or 88, or any of these sequences, wherein, independently, any amino acids, preferably no more than 20 amino acids, more preferably no more than 15, more preferably no more than 10, more preferably no more than 5, more preferably no more than 4, more preferably no more than 3, more preferably no more than 2, most preferably no more than 1 are substituted, preferably conservatively substituted, according to Table 3.

In a most preferred embodiment of the invention, the second binding moiety comprises a sequence selected from any of SEQ ID NOs: 59-72, 86 or 88, or any of these sequences, wherein, independently, any amino acids, preferably no more than 20 amino acids, more preferably no more than 15, more preferably no more than 10, more preferably no more than 5, more preferably no more than 4, more preferably no more than 3, more preferably no more than 2, most preferably no more than 1 are substituted, preferably conservatively substituted, according to Table 3.

In another preferred embodiment of the invention, the first binding fragment comprises the sequence set forth in SEQ ID NO: 85 and the second binding fragment comprises the sequence set forth in SEQ ID NO: 86.

In another preferred embodiment of the invention, the first binding fragment comprises the sequence set forth in SEQ ID NO: 85 and the second binding fragment comprises the sequence set forth in SEQ ID NO: 88.

In another preferred embodiment of the invention, the binding molecule comprises or consists of the sequence indicated in SEQ ID NO: 87.

The CDR1, CDR2 and CDR3 sequences or framework regions may be exchanged between species. For example, CDR sequences from a llama immunoglobulin molecule may be selected and exchanged with CDR sequences from a human immunoglobulin molecule to produce a human immunoglobulin molecule having the specificity derived from the llama CDR sequences. This may be advantageous because the human sequence may be less immunogenic to humans than an antibody comprising the original llama framework sequence. Such an exchange of sequences is known as humanization. Thus, the immunoglobulin molecules of the invention may have human-derived immunoglobulin sequences or immunoglobulin sequences derived from other animals such as, but not limited to, a camel, a llama, a shark, and have the CDR1, CDR2 and CDR3 sequences replaced by the CDR sequences of the invention to provide binding to human CD1d. In other words, a binding molecule of the invention may comprise a humanized single-domain antibody with the CDRs described herein. For example, a single-domain antibody may have human framework sequences and the CDRs described herein.

Thus, the invention provides a binding molecule that provides activation of iNKT cells and at the same time reduces the activation of Vδ1+ T cells. In a preferred embodiment, there is provided a binding molecule according to the invention or a binding molecule for use according to the invention, wherein the binding molecule also comprises a tumor-targeting moiety. The tumor-targeting moiety comprises a binding moiety capable of specifically binding to a tumor antigen. Preferably, the binding molecule also comprises a binding moiety that is capable of binding Vγ9Vδ2-TCR and is capable of competing for binding to Vγ9Vδ2-TCR with any of the VHHs shown in Table 1.

Tumor antigens are proteins produced by tumor cells that induce an immune response, in particular T-cell mediated immune responses. The choice of an antigen-binding fragment of the invention will depend on the specific type of cancer being treated. Tumor antigens are well known in the art and include, for example, glioma-associated antigen, carcinoembryonic antigen (CEA), EGFRvIII, interleukin-11 receptor alpha (IL-1 1Ra), interleukin-13 receptor alpha-2 subunit (IL-13Ra or CD213A2), epidermal growth factor receptor (EgFr), B7H3 (CD276), Kit (CD117), carbonic anhydrase (CA-IX), CS-1 (also called CD2 subset 1), cell surface-associated mucin 1 (MUC1), BCMA, an oncogene fusion protein consisting of a breakpoint cluster region (BCR) and Abelson murine leukemia viral oncogene homolog 1 (Abl) bcr-abl, ERBB2 receptor tyrosine protein kinase (HER2/neu), β-human chorionic gonadotropin, alpha-fetoprotein (AFP), anaplastic lymphoma kinase (ALK), CD19, CD123, cyclin B1, lectin-reactive AFP, Fos-related antigen 1, adrenoreceptor

- 22 048784 receptor beta 3 (ADRB3), thyroglobulin, tyrosinase; ephrin receptor type A 2 (EphA2), receptor for advanced glycation end products (RAGE-1), renal ubiquitin protein 1 (RU1), renal ubiquitin protein 2 (RU2), synovial sarcoma X breakpoint 2 (SSX2), anchor protein kinase 4 (AKAP-4), lymphocyte-specific protein tyrosine kinase (LCK), proacrosin-binding protein sp32 (OY-TES1), Pax-related box protein 5 (PAX5), squamous cell carcinoma T-cell recognition antigen 3 (SART3), C lectin-like molecule type 1 (CLL-1 or CLECL1), fucosyl GM1, hexasaccharide moiety of glycoceramide globoH (GloboH), MN-CA IX, epithelial cell adhesion molecule (EPCAM), EVT6-AML, transglutaminase 5 (TGS5), human telomerase reverse transcriptase (hTERT), polysialic acid, placenta-specific protein 1 (PLAC1), intestinal carboxylesterase, LewisY antigen, Lewis sialyl adhesion molecule (sLe), lymphocyte antigen 6 complex, K locus 9 (LY6K), mutated heat shock protein 70-2 (mut hsp70-2), M-CSF, avian v-myc, neuroblastoma-derived myelocytomatosis viral oncogene homolog (MYCN), Ras homolog family member C (RhoC), tyrosinase-related protein 2 (TRP-2), cytochrome P450 1B1 (CYP1B1), CCCTC-binding factor (zinc finger protein)-like (BORIS or Brother regulator of imprinted sites, Brother of the Regulator of Imprinted Sites), prostase, prostate-specific antigen (PSA), Pax-3 related box protein (PAX3), prostatic acid phosphatase (PAP), cancer/testis antigen 1 (NY-ESO-1), cancer/testis antigen 2 (LAGE-1a), LMP2, neuronal cell adhesion molecule (NCAM), p53 tumor protein (p53), p53 mutant, rat sarcoma mutant (Ras), glycoprotein 100 (gp100), prostein, OR51E2, pannexin 3 (PANX3), prostate-specific membrane antigen (PSMA), prostate stem cell antigen (PSCA), high molecular weight melanoma-associated antigen (HMWMAA), hepatitis A virus cellular receptor 1 (HAVCRI), vascular endothelial growth factor receptor 2 (VEGFR2), platelet-derived growth factor beta (PDGFR-beta), legumain, human papillomavirus E6 (HPV E6), human papillomavirus E7 (HPV E7), survivin, telomerase, sperm protein 17 (SPA17), stage-specific embryonic antigen 4 (SSEA-4), tyrosinase TCR gamma alternative reading frame protein (TARP), Wilms tumor protein (WT1), prostate carcinoma tumor antigen 1 (PCT-1), melanoma apoptosis inhibitor (ML-IAP), MAGE, melanoma-associated antigen 1 (MAGE-A1), testicular melanoma antigen-1 (MAD-CT-1), testicular melanoma antigen-2 (MAD-CT-2), melanoma T-cell recognition antigen 1 (MelanA/MARTI), family X antigen, element 1A (XAGE1), elongation factor 2 mutated (ELF2M), ERG (TMPRSS2 ETS fusion gene), N-acetylglucosaminyltransferase V (NA17), neutrophil elastase, sarcoma translocation breakpoints, mammary differentiation antigen (NYBR-1), ephrin B2, CD20, CD22, CD24, CD30, CD33, CD38, CD44v6, CD97, CD171, CD179a, androgen receptor, insulin growth factor (IGF)-I, IGF-II, IGF-I receptor, ganglioside GD2 (GD2), o-acetyl-GD2 ganglioside (OAcGD2), ganglioside GD3 (aNeu5Ac(2-8)aNeu5Ac(2-3)bDGalp(1-4)bDGlcp(1-1)Cer), GM3 ganglioside (aNeu5Ac(2-3)bDGa3p(1-4)bDGlcp(1-1)Cer), G protein-coupled receptor class C group 5 element D (GPRC5D), G protein-coupled receptor 20 (GPR20), chromosome X open reading frame 61 (CXORF61), folate receptor (FRa), folate receptor beta, receptor tyrosine kinase-like orphan receptor 1 (ROR1), Fms-like tyrosine kinase 3 (Flt3), tumor-associated glycoprotein 72 (TAG72), Tn antigen (TN Ag or (GalNAca-Ser/Thr)), angiopoietin-binding cell surface receptor 2 (Tie 2), tumor endothelial marker 1 (TEM1 or CD248), tumor endothelial marker 7-related (TEM7R), claudin 6 (CLDN6), thyroid stimulating hormone receptor (TSHR), uroplakin 2 (UPK2), mesothelin, serine protease 21 (testisin or PRSS21), epidermal growth factor receptor (EGFR), fibroblast activation protein alpha (FAP), olfactory receptor 51E2 (OR51E2), ETS translocation variant gene 6 located on chromosome 12p (ETV6-AML), CD79a, CD79b, CD72, leukocyte-associated immunoglobulin-like receptor 1 (LAIR1), Fc fragment IgA receptor (FCAR or CD89), leukocyte immunoglobulin-like receptor subfamily A member 2 (LILRA2), CD300 family member f-like molecule (CD300LF), C-type lectin domain family 12 member A (CLEC12A), bone marrow stromal cell antigen 2 (BST2), EGF-like module-containing mucin-like hormone receptor 2-like (EMR2), lymphocyte antigen 75 (LY75), glypican-3 (GPC3), Fc receptor-like 5 (FCRL5), immunoglobulin lambda-like polypeptide 1 (IGLL1), folate receptor (FRa), mesothelin, EGFR variant III (EGFRvIII), B-cell maturation antigen (BCMA), GD2, CLL-1, CA-IX, MUC1, HER2, and any combination thereof. In one embodiment, the tumor-targeting moiety is an immunoglobulin that specifically binds to PDL1, EGFR, CD40, Her2, PSMA, MUC-1, CEA, c-met, CD19, CD20, BCMA, Her3, AFP, CAIX, or CD38.

As noted, the binding molecule according to the invention provides for the creation of a microenvironment that is useful for the destruction of tumor cells by, for example, iNKT cells and Yγ9Yδ2 T cells. Thus, a binding molecule comprising a first binding moiety and a second binding moiety according to the invention is provided for use in the treatment of a tumor. Such a binding molecule is not only effective against CD1d+ tumors, but also against tumors that do not express CD1d per se, but (partially) express sex

- 23 048784 target CD1d+ suppressor cells (e.g. MDSC or tumor-associated macrophages (TAM)) in the tumor environment. Preferably, the tumor is selected from hematological malignancies such as T-cell lymphoma, multiple myeloma, acute myeloid leukemia, acute lymphoblastic leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, mantle cell lymphoma, B-cell lymphoma, smoldering myeloma, non-Hodgkin's lymphoma, Hodgkin's lymphoma, myelomonocytic leukemias, lymphoplasmacytic lymphoma, hairy cell leukemia, splenic marginal zone lymphoma, or solid tumors such as renal cell carcinoma, melanoma, colorectal cancer, head and neck cancer, breast cancer, prostate cancer, lung cancer, pancreatic cancer, esophageal cancer, small bowel cancer, central nervous system tumors, medulloblastomas, hepatocellular carcinoma, cancer ovarian, glioma, neuroblastoma, urothelial carcinomas, bladder cancer, sarcoma, penile cancer, basal cell carcinoma, Merkel cell carcinoma, neuroendocrine carcinoma, neuroendocrine tumors, carcinoma of unknown primary (CUP), thymoma, vulvar cancer, cervical cancer, testicular cancer, cholangiocarcinoma, appendiceal carcinoma, mesothelioma, ampullary carcinoma, anal cancer, and choriocarcinoma.

Pharmaceutical compositions, dosages, routes of administration and methods of treatment.

In another main aspect, the invention relates to a pharmaceutical composition comprising a binding molecule, such as an antibody, comprising a first binding moiety that is capable of competing with 1D12 for binding to a CD1d molecule and comprising a second binding moiety that is capable of specifically binding to a Vγ9Vδ2 TCR, wherein the binding molecule is capable of activating Vγ9Vδ2 T cells, and a pharmaceutically acceptable excipient.

Polypeptides such as antibodies can be combined with pharmaceutically acceptable carriers or diluents, as well as any other known adjuvants and excipients, according to standard methods, such as those described in (Rowe et al., Handbook of Pharmaceutical Excipients, 2012 June, ISBN 9780857110275). Pharmaceutically acceptable carriers or diluents, as well as any other known adjuvants and excipients, must be suitable for the polypeptides or antibodies and the selected route of administration. Suitable for carriers and other components of pharmaceutical compositions is determined based on the absence of a significant adverse effect on the desired biological properties of the selected compound or pharmaceutical composition of the present invention (e.g., less than a significant effect (10% or less relative inhibition, 5% or less relative inhibition, etc.) in binding to the antigen). The pharmaceutical composition may also include diluents, fillers, salts, buffers, detergents (e.g., nonionic detergents such as Tween-20 or Tween-60), stabilizers (e.g., sugars or amino acids without proteins), preservatives, tissue fixatives, solubilizers, and/or other materials suitable for inclusion in the pharmaceutical composition. Other pharmaceutically acceptable excipients and carriers include any and all suitable solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents, antioxidants, and absorption delaying agents, and the like, that are physiologically compatible with the binding molecule of the present invention.

The invention provides methods for treating a disease or disorder comprising administering a binding molecule as described herein to a subject in need thereof. In one embodiment, the subject is a human. The method of the invention comprises administering an effective amount of binding molecules. The term treatment means administering an effective amount of a therapeutically active polypeptide of the present invention to alleviate, improve, arrest or eliminate (cure) symptoms or disease states. The term effective amount or therapeutically effective amount means an amount that is effective, at the required dosages and for the periods of time, to achieve the desired therapeutic result. A therapeutically effective amount of a polypeptide, such as an antibody, may vary according to factors such as the stage of the disease, the age, sex and body weight of the subject, and the ability of the antibody to elicit the desired response in the subject. A therapeutically effective amount is also an amount in which any toxic or deleterious effects of the antibody or antibody portion are outweighed by the therapeutically beneficial effects.

Administration may be by any suitable route, but will typically be parenteral, such as intravenous, intramuscular, or subcutaneous. Effective dosages and dosing routes for a binding molecule, such as an antibody, depend on the disease or condition being treated and can be determined by those skilled in the art.

Binding molecules, such as antibodies of the present invention, may also be administered as part of a combination therapy, i.e., may be combined with other therapeutic agents that are relevant to the disease or condition being treated. Accordingly, in one embodiment, the antibody-containing medicament is intended to be combined with one or more other therapeutic agents, such as cytotoxic

- 24 048784 therapeutic, chemotherapeutic or antiangiogenic agents. Such combined administration may be simultaneous, separate or sequential. Thus, in another embodiment, the present invention provides a method for treating or preventing a disease, such as cancer, the method comprising administering to a subject in need thereof a therapeutically effective amount of a binding molecule or pharmaceutical composition of the present invention in combination with radiotherapy and/or surgery.

Other aspects and embodiments of the invention.

In another aspect, the invention relates to a binding molecule comprising a first binding moiety that is capable of specifically binding to a CD1d molecule, wherein the binding molecule is capable of reducing the activation of V51 T cells and is capable of activating iNKT cells, for use in the treatment of diseases caused, maintained and/or stimulated by the activation of CD1d-restricted Vδ1+ T cells, preferably for use in the treatment of peripheral CD1d-restricted Vδ1+ T cell lymphoma.

In another aspect, the invention provides a binding molecule comprising a first binding moiety that is capable of specifically binding to a CD1d molecule and comprising a second binding moiety that is capable of specifically binding to a Vγ9Vδ2 TCR, wherein the binding molecule is capable of reducing activation of Vδ 1 T cells, is capable of activating iNKT cells, and is capable of activating Vγ9Vδ2 T cells.

In another aspect, the invention relates to an antibody comprising a first binding moiety that is capable of competing with the single domain antibody 1D12 for binding to a CD1d molecule, for use in the treatment of diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells, preferably for use in the treatment of peripheral CD1d-restricted Vδ1+ T cell lymphoma. Preferably, the antibody is capable of reducing the activation of Vδ 1 T cells and/or is capable of activating iNKT cells.

In another aspect, the invention relates to an antibody, such as a bispecific antibody, comprising a first binding fragment that is capable of competing with 1D12 for binding to a CD1d molecule and comprising a second binding fragment that is capable of specifically binding to a Vγ9Vδ2 TCR, wherein the binding molecule is capable of activating Vγ9Vδ2 T cells. Preferably, the antibody is capable of reducing the activation of V51 T cells and/or is capable of activating iNKT cells. Preferably, the first and/or second binding fragment is a single-domain antibody.

Methods for producing binding molecules according to the invention.

The binding molecules of the invention, such as polypeptides, in particular antibodies, are typically produced by recombinant methods, i.e. by expressing nucleic acid constructs encoding the polypeptides in suitable host cells, followed by recovery of the resulting recombinant polypeptide from cell culture. The nucleic acid constructs can be prepared using standard molecular biology techniques that are well known in the art. The constructs are typically introduced into the host cell using a vector. Suitable nucleic acid constructs and vectors are well known in the art. Host cells suitable for recombinant expression of polypeptides, such as antibodies, are well known in the art and include CHO, HEK-293, Expi293F, PER-C6, NS/0 and Sp2/0 cells.

Examples

Materials.

Cell lines.

Stably transduced CD1d human Epstein-Barr virus-transformed B lymphoblast cell line C1R and human JY cell line were grown in Dulbecco's modified Iscove medium (Cat. No. 12-722F; Lonza, Basel, Switzerland) supplemented with 10% (v/v) fetal bovine serum (Cat. No. SV30160.03; HyClone GE Healthcare, Chalfont, St Giles, UK), 0.05 mM β-mercaptoethanol, 100 IU/mL penicillin sodium, 100 μg/mL streptomycin sulfate, and 2.0 mM L-glutamine (Cat. No. 10378-016; Life Technologies, Carlsbad, CA). The stably transduced CD1d human cervical adenocarcinoma HeLa cell line was cultured in Dulbecco's modified Eagle's medium (Cat. No. BE12-709F; Lonza) supplemented with 10% (v/v) fetal calf serum, 0.05 mM β-mercaptoethanol, 100 IU/ml penicillin sodium, 100 μg/ml streptomycin sulfate, and 2.0 mM L-glutamine. The human myeloma cell line MM.1s with/without mcherry/luc and the stably transduced CD1d human acute T-lymphoblastic leukemia cell line CCRF-CEM, the V51 sulfatide-CD1d-restricted TCR-transduced human acute T-cell leukemia cell line Jurkat and the human acute myeloid leukemia cell lines M0LM-13 and N0M0-1 were cultured in RPMI-1640 medium (cat. no. BE12-115F; Lonza) supplemented with 10% (v/v) fetal calf serum, 0.05 mM β-mercaptoethanol, 100 IU/ml penicillin sodium, 100 μg/ml streptomycin sulfate and 2.0 mM L-glutamine. The genetic characteristics of CCRF-CEM and MM1.s were determined using single-locus PCR technology and found to be identical to published DNA profiles.

- 25 048784 cells were negative for mycoplasma when tested and were subjected to frequent purity testing (transfectants) by flow cytometry.

Flow cytometry and monoclonal antibodies.

The following antibodies were used in this study: fluorescein isothiocyanate (FITC)-conjugated Vδ2, FITC CD69, phycoerythrin (PE)-allophycocyanin (APC)-conjugated CD25 (Cat. #555432 and #340907), and APC CD3 were purchased from BD Biosciences (Franklin Lakes, NJ). Phycoerythrin-cyanine 7-conjugated Va24 (Cat. #PN A66907) and Ve 11 PE (Cat. #IM2290) were purchased from Beckman Coulter (Brea, CA). 7-Aminoactinomycin D (7-AAD) was purchased from Sigma (St Louis, MO), PE Vy9 from Biolegend (San Diego, USA), PE CD107a from Miltenyi (Miltenyi Biotec, Bergisch Gladbach, Germany), and FITC annexin V from VPS Diagnostics (Hoever, the Netherlands) (cat. no. A700). Tetramers were prepared in-house. Flow cytometric staining was performed in FACS buffer (PBS supplemented with 0.1% BSA and 0.02% sodium azide) for 30 min at 4°C unless otherwise stated. Samples were analyzed on a FACS Fortessa (BD Biosciences). Generation of DC, iNKT and Vy9VS2-T cell lines moDC and primary human iNKT cells and γδ T cells were generated as described previously (De Bruin et al. (2016) Clin Immunol 169:128). Briefly, monocytes were isolated from peripheral blood mononuclear cells using CD14 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) and cultured in complete RPMI-1640 medium in the presence of 1000 U/mL granulocyte-macrophage colony-stimulating factor (Sanofi Leukine, Bridgewater, NJ) and 20 ng/mL IL-4 (Cat. No. 204-IL/CF; R&D Systems, Minneapolis, MN) for 5–7 days and further matured with 100 ng/mL lipopolysaccharide (LPS) (Cat. No. L6529; Sigma) in the presence or absence of 100 ng/mL α-GalCer (Cat. No. KRN7000; Funakoshi, Tokyo, Japan) for 48–72 h. iNKT cells were isolated from healthy volunteer peripheral blood cells using magnetic bead sorting and stimulated weekly with mature a-GalCer-loaded moDCs in Issel medium supplemented with 1% human AB serum, 10 U/ml IL-7 (cat. #207-IL/CF; R&D Systems), and 10 ng/ml IL-15 (cat. #34-8159; eBioscience). γδT cells were isolated from peripheral blood mononuclear cells of healthy volunteers using magnetic bead sorting and stimulated weekly with pamidronate-loaded (10 μM) moDC (PCH, Pharmachemie BV, Haarlem, The Netherlands) in Issel medium supplemented with 1% human serum AB, 100 U/ml IL-2 (BioVision, Mountain View, California, USA), 10 U/ml IL-7, and 10 ng/ml IL-15. Alternatively, γδ T cells were stimulated weekly with irradiated feeder cells (1 x 10 ⁶ mixed PBMCs from two donors and 0.1 x 10 ⁶ JY cells), 10 IU/ml rhIL-7, 10 μg/ml rhIL-15, and 50 ng/ml PHA in RPMI-1640 medium supplemented as described above. Depending on the culture density, cultured cells were split and fresh culture medium was added. Pure (>95% Va24+ Υβ11+ or Vy9+ Vδ2+) iNKT and γδT cells were used for experiments. Generation of anti-CD1d- and anti-γδTCR-specific VHH.

Anti-CD1d- and anti-γδTCR-specific VHH were identified and prepared as described previously (Lameris R. et al. (2016) Immunology 149(1):111; De Bruin et al. (2016) Clin Immunol 169:128). Unlabeled 1D12, 1D22, and 1D12-5C8 were obtained from UPE (Utrecht, the Netherlands).

In vivo xenograft mouse model of multiple myeloma (MM). The disseminated MM model was generated by intravenous transfer of CD1d ⁺ MM cells into NOD scid gamma (NSG) mice. Female NSG mice (Charles River) aged 18–26 weeks were irradiated with 2 Gy for 24 h before intravenous (i.v.) injection of 2.5 x 10 ⁶ MM.1s.mcherry/luc. CD1d cells into the tail vein (day 0). On days 7, 14, and 21, 1 x 10 ⁷ human iNKT cells, human γδ T cells, or a mixture of both (in a 1:1 ratio) were injected intravenously. Mice were injected intraperitoneally (i.p.) with PBS or the bispecific antibody 1D12-5C8 (100 μg/mouse) twice weekly. Mice were sacrificed when predefined human endpoints were reached. Animal experiments were approved by the Dutch Central Authority for Scientific Procedures on Animals (CCD).

Example 1. Modulation of iNKT cell activation.

To assess the ability of 1D12 and 1D22 to stimulate or inhibit iNKT cell activation, 5 × 10 ⁴ Hela-CD1d cells were seeded into wells of a 96-well culture plate and pulsed overnight with vehicle control (DMSO 0.01%) or 100 ng/ml α-GalCer. Cells were then washed with PBS and incubated with medium or anti-CD1d-specific VHH for one hour at the indicated concentrations. Next, 5 × 104 pure and stable iNKT were added to each well. After 24 h, culture supernatants were analyzed for (stimulation or inhibition of) cytokine production using CBA (BD Biosciences), while iNKT cells were harvested and analyzed for CD25 expression by flow cytometry. As shown in Fig. 1, we identified an anti-CD1d VHH (clone 1D22) that completely blocked iNKT cell activation and cytokine production (P<0.0001) and thus recognition of the CD1d-a-GalCer complex. In contrast, anti-CD1d VHH clone 1D12 was found to stimulate activation of CD1d-restricted iNKT cells even in the absence of exogenously added glycolipid Ag (Fig. 1A and C) (P<0.0001).

- 26 048784

Example 2. Modulation of Jurkat-V51 cell activation.

It is known that iNKT cells dock in the region of the extreme F' pocket of CD1d, in contrast to sulfatide-CD1d-restricted Vδ1 T cells, which dock closer to the A' pocket. Therefore, we analyzed the effect of 1D12 and 1D22 on sulfatide-CD1d-restricted Vδ1 T cells. To assess the effect of 1D12 and 1D22 on Jurkat-Vδ1 cell activation, 1 x 10 ⁵ C1R-CD1d cells were seeded in wells of a 96-well culture plate and pulsed for 2 h with vehicle control (DMSO 0.05%) or 25 μg/ml sulfatide. Cells were then incubated with culture medium or anti-CD1d-specific VHH for 1 h at 100 nM (not shown) or 1000 nM. Then, 5 x ¹⁰⁴ Jurkat-Vδ1 were added to each well. After 24 h, Jurkat-Vδ1 cells were harvested and analyzed for CD69 expression by flow cytometry. As shown in Fig. 2B, addition of 1D12 during co-culture completely abolished Vδ1-Jurkat activation, while 1D22 had only a limited effect on expression of the activation marker CD69. Incubation with 100 nM or 1000 nM resulted in similar results (data not shown).

To assess the effect of 1D12 and 1D22 on CD1d-tetramer binding on Jurkat-Vδ1 cells, endogenous or sulfatide-loaded CD1d-PE tetramers were incubated with PBS (control), 1D12, or 1D22 (VHH:CD1d ratio ~4:1) for 30 min at room temperature, after which the tetramers were added to Jurkat-V1 cells (final tetramer concentration 2 μg/ml) in combination with CD3-APC and incubated for 45 min at 4 °C. Data were analyzed by flow cytometry. Incubation of sulfatide-loaded CD1d tetramers with 1D12 completely prevented binding to Jurkat-V1 cells, whereas 1D12 had only a limited effect (Fig. 2A). These data support the ability of anti-CD1d VHH to modulate the reactivity of specific CD1d-restricted T cells, which is in contrast to known mAbs such as 51.1 mAb, which blocks CD1d-TCR interactions of a broad range of CD1d-restricted T cells (Nambiar et al. (2015) MAbs 7:638; Migalovich Sheikhet et al. (2018) Front Immunol 9:753).

Example 3. Dual activation of iNKT and Vγ9Vδ2 T cells by bispecific anti-CD1d-anti-Vγ9VS2 TCR VHH.

Previously well-characterized anti-Vγ9Vδ2-TCR VHHs were fused to tumor-associated antigen-specific VHHs for anti-cancer therapeutic purposes. CD1d is expressed in various (hematological) diseases and on tumor-associated macrophages and myeloid-derived suppressor cells and may thus be used as an anti-cancer therapeutic target. To assess the ability of 1D12-5C8 to induce dual activation of iNKT and Vγ9Vδ2 T cells leading to target tumor lysis, ^1x105 CCRF-CEM cells were incubated with ^5x104 iNKT cells, ^5x104 Vγ9Vδ2 T cells, or ^5x104 iNKT/Vγ9Vδ2 T mixtures (1:1 ratio) in the presence of medium alone, monovalent 1D12, or bispecific 1D12-5C8. After 4 h, effector cell degranulation was measured based on CD107a expression and analyzed by flow cytometry. To analyze cytotoxicity towards target cells, live CCRF-CEM cells (Annexation V- and 7-AAD-negative) were quantified after 16 h of co-culture using cell counting beads in a flow cytometric assay.

To determine the ability of 1D12-5C8 to support iNKT and Vγ9Vδ2-T expansion and control tumor growth, freshly isolated iNKT and Vγ9Vδ2-T from the same donor were cultured for 1 week. Then, ^5x104 MM.1s-CD1d cells were incubated with medium or 1D12-5C8 (50 nM) for 30 min, after which iNKT, Vγ9Vδ2-T cells, or their mixture (2:3 ratio) were added at a target:effector ratio of 10:1. Live MM.1s-CD1d (or MOLM-13 or NOMO-1), iNKT, and Vγ9Vδ2 T cells (7AAD-negative) were quantified after 7 days using flow cytometric cell counting beads.

As shown in Fig. 3A, robust simultaneous degranulation of iNKT and Vγ9Vδ2 T cells was observed only in the presence of 1D12-5C8. Moreover, activation of effector cells resulted in a strong reduction in the number of viable tumor cells (Fig. 3B).

An unfavorable effector:target ratio in vivo typically requires expansion of tumor-targeted effector cells to control tumor growth. To analyze whether the bispecific 1D12-5C8 VHH could induce both effector cell expansion and tumor control under these conditions, MM.1s-CD1d cells were incubated with 1D12-5C8, after which iNKT, Vγ9Vδ2 T cells, or a mixture of both were added at an effector:target ratio of 1:10. The ability of 1D12-5C8 to induce tumor expansion and control tumor growth was assessed after 7 days of co-culture using quantitative flow cytometry-based analysis of the cells. As shown in Fig. 4A, iNKT expansion was observed in the presence of the bispecific construct. However, Vγ9Vδ2 T cells only showed expansion in the presence of both iNKT cells and the bispecific construct. Moreover, robust tumor growth control was induced by the bispecific construct in combination with effector cells (Fig. 4B). Similarly, tumor growth control was

- 27 048784 choline and effector cell expansion were observed in the case of acute myeloid leukemia cell lines MOLM-13 and NOMO-1, indicating high antitumor efficacy and broad applicability of this bispecific anti-CD1d-anti-Vγ9Vδ2-TCRVHH.

Example 4. Competition for binding of 1D12 and 1D12-5C8.

To assess the ability of 1D12 binding to interfere with 1D12-5C8 binding, 1 x 10 ⁵ MM1sCD1d cells were incubated with PBS (negative control, NC), 1D12 (1000 nM), 1D22 (1000 nM), or anti-CD1d mAb 51.1 (100 nM) for 45 min, followed by the addition of PBS (NC) or NHS-biotin (ThermoFischer Scientific Inc., Waltham, MA)-coupled 1D12-5C8 (100 nM) for an additional 30 min at 4 °C. After extensive washing, samples were stained with APC streptavidin (eBioscience, San Diego, CA) and analyzed by flow cytometry.

To assess competition between 1D12 and 1D12-5C8, CD1d-expressing MM cells were further incubated with PBS, 1D12, 1D22 (which should not interfere with 1D12-5C8 binding), or anti-CD1d mAb 51.1) followed by biotinylation of 1D12-5C8. The ability of 1D12-5C8 to form a complex with CD1d was then determined by flow cytometric streptavidin-APC binding assay. As shown in Fig. 5, preincubation with 1D12 or anti-CD1d mAb 51.1, but not 1D22, significantly reduced 1D12-5C8 binding.

Example 5. In vivo murine xenograft model of multiple myeloma (MM). The antitumor efficacy of the bispecific CD1d/Vδ2-binding antibody 1D12-5C8 was analyzed in an in vivo model in which mice were inoculated i.v. with MM.1s.mCherry/luc.CD1d cells to generate a disseminated MM model followed by three i.v. infusions of human iNKT cells, human γδ T cells, or a mixture of both 7 days after tumor inoculation with/without the combination of 1D12-5C8. While 1D12-5C8 alone twice weekly had no effect (median survival 47 days vs 49.5 days, P>0.05), combination treatment with 1D12-5C8 and iNKT cells significantly (p<0.0001) prolonged survival compared with iNKT alone (median survival 58.5 days), with all mice alive at the end of the study (day 90). Compared with human γδ T cell treatment alone, treatment with human γδ T cells and 1D12-5C8 showed a trend to increase median survival from 48 days to 60 days (p=0.16). Infusion of both iNKT cells and γδ T cells with i.p. 1D12-5C8 twice weekly significantly increased survival, with 7/8 mice alive at the end of the study (day 90) (p>0.0001) compared to a mixture of these cells alone without antibody (median survival 55 days).

Claims (8)

1. A binding molecule comprising: (i) a first binding fragment of an antibody that is capable of specifically binding to CD1d and comprises SEQ ID NO: 85, and (ii) a second binding fragment that is capable of specifically binding to Vγ9Vδ2-TCR and comprises SEQ ID NO: 86, wherein the binding molecule is capable of activating and inducing expansion of iNKT cells and Vγ9Vδ2 T cells. 2. The binding molecule according to claim 1, characterized in that the binding molecule is capable of reducing the activation of V51 T cells. 3. A binding molecule according to any of the preceding paragraphs, characterized in that the first and/or second binding fragment is a single-domain antibody. 4. A binding molecule according to any of the preceding claims, characterized in that the binding molecule comprises the sequence indicated in SEQ ID NO: 87. 5. Use of a binding molecule according to any of the preceding paragraphs for the treatment of diseases caused, maintained and/or promoted by activation of CD1d-restricted Vδ1+ T cells. 6. Use of a binding molecule according to any of the preceding paragraphs in the treatment of a tumor expressing CD1d. 7. The use according to claim 6, wherein the tumor is selected from hematological malignancies such as T-cell lymphoma, multiple myeloma, acute myeloid leukemia, acute lymphoblastic leukemia, chronic lymphocytic leukemia, mantle cell lymphoma, B-cell lymphoma, smoldering myeloma, Hodgkin's lymphoma, myelomonocytic leukemias, lymphoplasmacytic lymphoma, hairy cell leukemia, splenic marginal zone lymphoma, or solid tumors such as renal cell carcinoma, melanoma, colorectal cancer, head and neck cancer, breast cancer, prostate cancer, lung cancer, pancreatic cancer, esophageal cancer, small bowel cancer, central nervous system tumors, medulloblastomas, hepatocellular carcinoma, ovarian cancer, glioma, neuroblastoma, urothelial carcinomas, bladder cancer, sarcoma, penile cancer, basal cell carcinoma, Merkel cell carcinoma, neuroendocrine carcinoma, neuroendocrine tumors, carcinoma of unknown primary (CUP), thymoma, vulvar cancer, cervical cancer, testicular cancer, - 28 048784 cholangiocarcinoma, appendiceal carcinoma, mesothelioma, ampullary carcinoma, anal cancer, and choriocarcinoma. 8. A pharmaceutical composition comprising a binding molecule according to claim 1 and a pharmaceutically acceptable excipient.